Nothing Special   »   [go: up one dir, main page]
Next Issue
Volume 21, February-2
Previous Issue
Volume 21, January-2
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
7.3
3.4
Journals
Sensors
Volume 21
Issue 3
sensors-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Sensors, Volume 21, Issue 3 (February-1 2021) – 347 articles

Cover Story (view full-size image): In the last decade, Wake-up Radio technology has gained more and more importance for IoT applications. This technology is based on an ultra-low-power, always-listening radio receiver that prevents the main IoT transceiver to continuously listen to the channel, lowering its power consumption. On the other hand, LoRa is one of the emerging long-range standards that facilitate the environmental monitoring of wide areas, but generally suffer from high latency in downlink. This paper proposes an original combination of Lora technology and Wake-up radio to improve the performance (latency and energy) of the LoRaWAN downlink transmission. It is also shown that this combination is particularly suitable to nodes that have the capability to harvest surrounding energy. This research is mainly funded by French ANR and takes place in the collaborative project “Wake-Up”. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
19 pages, 10620 KiB  
Article
Characterisation of Textile Embedded Electrodes for Use in a Neonatal Smart Mattress Electrocardiography System
by Henry Dore, Rodrigo Aviles-Espinosa, Zhenhua Luo, Oana Anton, Heike Rabe and Elizabeth Rendon-Morales
Sensors 2021, 21(3), 999; https://doi.org/10.3390/s21030999 - 2 Feb 2021
Cited by 7 | Viewed by 4661
Abstract
Heart rate monitoring is the predominant quantitative health indicator of a newborn in the delivery room. A rapid and accurate heart rate measurement is vital during the first minutes after birth. Clinical recommendations suggest that electrocardiogram (ECG) monitoring should be widely adopted in [...] Read more.
Heart rate monitoring is the predominant quantitative health indicator of a newborn in the delivery room. A rapid and accurate heart rate measurement is vital during the first minutes after birth. Clinical recommendations suggest that electrocardiogram (ECG) monitoring should be widely adopted in the neonatal intensive care unit to reduce infant mortality and improve long term health outcomes in births that require intervention. Novel non-contact electrocardiogram sensors can reduce the time from birth to heart rate reading as well as providing unobtrusive and continuous monitoring during intervention. In this work we report the design and development of a solution to provide high resolution, real time electrocardiogram data to the clinicians within the delivery room using non-contact electric potential sensors embedded in a neonatal intensive care unit mattress. A real-time high-resolution electrocardiogram acquisition solution based on a low power embedded system was developed and textile embedded electrodes were fabricated and characterised. Proof of concept tests were carried out on simulated and human cardiac signals, producing electrocardiograms suitable for the calculation of heart rate having an accuracy within ±1 beat per minute using a test ECG signal, ECG recordings from a human volunteer with a correlation coefficient of ~ 87% proved accurate beat to beat morphology reproduction of the waveform without morphological alterations and a time from application to heart rate display below 6 s. This provides evidence that flexible non-contact textile-based electrodes can be embedded in wearable devices for assisting births through heart rate monitoring and serves as a proof of concept for a complete neonate electrocardiogram monitoring system. Full article
(This article belongs to the Special Issue ECG Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Delivery room mattress with embedded sensors.</p> Full article ">Figure 2
<p>Prototype system signal path with input, filtering and output stages identified.</p> Full article ">Figure 3
<p>(<b>a</b>) Exploded view of sensor: (1) deposited reference electrode material (conductive polymer ink or conductive textile fabric), (2) cotton substrate, (3) and (5) 3D printed housing, (4) EPS ASIC, (6) power and signal connections, (7) single strand copper connection to the sensor ground. (<b>b</b>) assembled EPS sensor, with stAandard copper electrode, without the cotton layer attachment.</p> Full article ">Figure 4
<p>Textile based EPS electrodes, silver conductive ink (<b>a</b>) and conductive textile (<b>b</b>).</p> Full article ">Figure 5
<p>(<b>a</b>) Skin-electrode interface and tissue model for impedance measurement setup, (<b>b</b>) Measured impedance of the skin electrode interface, (<b>c</b>) impedance mismatch between electrode pairs.</p> Full article ">Figure 6
<p>Filtering stage block diagram.</p> Full article ">Figure 7
<p>(<b>a</b>) Frequency response analysis of hardware and software stages, (<b>b</b>) power spectral density showing the combined effect of hardware and software filtering on a noisy test ECG signal injected directly into the front end.</p> Full article ">Figure 8
<p>Screenshot of prototype graphical user interface.</p> Full article ">Figure 9
<p>Experimental set up—electrode placement and interference layers for simulated and human ECG readings, left neonate mannequin, right human volunteer.</p> Full article ">Figure 10
<p>Photograph of prototype system hardware in neonate simulation environment: (a) front end and GUI (Dimensions 195 × 100 × 90mm); (b) USB 5V power bank; (c) EPS sensors and textile based electrodes; (d) digital to analogue converter for test signal generation; (e) neonate mannequin with internal antenna; (f) commercial delivery room mattress.</p> Full article ">Figure 11
<p>5 s segments of a simulated ECG signal for (<b>a</b>) baseline copper electrodes; (<b>b</b>) conductive polymer ink; (<b>c</b>) conductive textile fabric.</p> Full article ">Figure 12
<p>(<b>a</b>) Single beat waveform and (<b>b</b>) Power spectral density of a simulated ECG simulated from the test mannequin for each electrode configuration.</p> Full article ">Figure 13
<p>15 s segments of a human signal ECG signal recorded with (<b>a</b>) baseline copper electrodes; (<b>b</b>) conductive polymer ink; (<b>c</b>) conductive textile fabric.</p> Full article ">Figure 14
<p>(<b>a</b>) Single beat waveform and (<b>b</b>) PSD of a human ECG for each electrode configuration.</p> Full article ">Figure 15
<p>Waveform averaging of the human ECG for (<b>a</b>) baseline copper electrodes; (<b>b</b>) conductive polymer ink; (<b>c</b>) conductive textile fabric.</p> Full article ">Figure 15 Cont.
<p>Waveform averaging of the human ECG for (<b>a</b>) baseline copper electrodes; (<b>b</b>) conductive polymer ink; (<b>c</b>) conductive textile fabric.</p> Full article ">Figure 16
<p>Wavelet delineation of a 7 s sample of recorded ECG, showing the location and duration of the waveform features.</p> Full article ">Figure 17
<p>mean P QRS and T lengths and RR intervals for each electrode case.</p> Full article ">Figure 18
<p>Connection event showing time from sensor application to first HR measurement.</p> Full article ">
16 pages, 13214 KiB  
Article
Signal Diversity for Laser-Doppler Vibrometers with Raw-Signal Combination
by Marvin Schewe and Christian Rembe
Sensors 2021, 21(3), 998; https://doi.org/10.3390/s21030998 - 2 Feb 2021
Cited by 9 | Viewed by 3674
Abstract
The intensity of the reflected measuring beam is greatly reduced for laser-Doppler vibrometer (LDV) measurements on rough surfaces since a considerable part of the light is scattered and cannot reach the photodetector (laser speckle effect). The low intensity of the reflected laser beam [...] Read more.
The intensity of the reflected measuring beam is greatly reduced for laser-Doppler vibrometer (LDV) measurements on rough surfaces since a considerable part of the light is scattered and cannot reach the photodetector (laser speckle effect). The low intensity of the reflected laser beam leads to a so-called signal dropout, which manifests as noise peaks in the demodulated velocity signal. In such cases, no light reaches the detector at a specific time and, therefore, no signal can be detected. Consequently, the overall quality of the signal decreases significantly. In the literature, first attempts and a practical implementation to reduce this effect by signal diversity can be found. In this article, a practical implementation with four measuring heads of a Multipoint Vibrometer (MPV) and an evaluation and optimization of an algorithm from the literature is presented. The limitations of the algorithm, which combines velocity signals, are shown by evaluating our measurements. We present a modified algorithm, which generates a combined detector signal from the raw signals of the individual channels, reducing the mean noise level in our measurement by more than 10 dB. By comparing the results of our new algorithm with the algorithms of the state-of-the-art, we can show an improvement of the noise reduction with our approach. Full article
(This article belongs to the Special Issue Laser Doppler Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Concept of the experiment to obtain velocity signals with signal dropouts; (<b>b</b>) Image of the actual test setup; laser beams go to the shaker in a straight line through the holes in the rotating disc.</p> Full article ">Figure 2
<p>(<b>a</b>) Concept of the second experimental setup with one active measuring head (CH4) and three passive measuring heads (CH1–CH3); (<b>b</b>) Image of the actual test setup measuring on a speaker.</p> Full article ">Figure 3
<p>Velocity signals of the individual channels (CH1–CH4) and the combined velocity signal (VeloComb.); (RBW = 2.5 Hz); additional detailed view of a small section to better show the combined signal.</p> Full article ">Figure 4
<p>Frequency spectra of the velocity signals shown in <a href="#sensors-21-00998-f003" class="html-fig">Figure 3</a> (smoothed for better visualization); (RBW = 2.5 Hz).</p> Full article ">Figure 5
<p>Detailed view of the velocity signals of the individual channels (CH1–CH4) as well as the combined signal (VeloComb); (RBW = 2.5 Hz).</p> Full article ">Figure 6
<p>(<b>a</b>) Raw signals of the individual channels (CH1–CH4) at the same time segment as the velocity signals shown in <a href="#sensors-21-00998-f005" class="html-fig">Figure 5</a>; (<b>b</b>) Upper signal envelope of the raw signals.</p> Full article ">Figure 7
<p>Raw signals of the individual channels and the combined signal (VeloComb) derived from the individual channels and weighting factors from Equation (2).</p> Full article ">Figure 8
<p>Visualization of the process of digitizing and splitting the individual channels into blocks (the shown signal is chosen only for visualization with a low frequency; in the real signal there are several hundred periods in each block).</p> Full article ">Figure 9
<p>(<b>a</b>) Raw signals of the individual channels (CH1–CH4) without phase correction; (<b>b</b>) Raw signals with phase correction (samples with dotted lines are only shown for visualization of the shift).</p> Full article ">Figure 10
<p>Section of the combined, demodulated velocity signal before and after interpolation to correct discontinuities and error due to phase shifting.</p> Full article ">Figure 11
<p>Comparison of the demodulated velocity signals of the channels CH1–CH4 and the combined signals (from [<a href="#B21-sensors-21-00998" class="html-bibr">21</a>] “VeloComb” and from our algorithm “RawComb”); zoomed sections only for combined signals to illustrate less noise; (RBW = 2.5 Hz).</p> Full article ">Figure 12
<p>(<b>a</b>) Comparison of the demodulated displacement signals of the four channels CH1–CH4 as well as the combined signals derived from the old and new algorithms; (RBW = 2.5 Hz). (<b>b</b>) Magnified section of the combined signals from (<b>a</b>) to illustrate the difference.</p> Full article ">Figure 13
<p>(<b>a</b>) Frequency spectra of the velocity signals with a detailed view of the vibration frequency at 100 Hz; (RBW = 2.5 Hz). (<b>b</b>) Smoothed frequency spectra for the combined and one individual velocity signal with the amplitude in dB (0 dB = 1 m/s); same scale as in (<b>a</b>).</p> Full article ">Figure 14
<p>Comparison of the demodulated velocity signals of the four channels CH1–CH4 and the combined signal from the raw signals (VeloComb); (RBW = 1 Hz).</p> Full article ">Figure 15
<p>Comparison of the demodulated displacement signals of the four channels CH1–CH4 and the combined signal; (RBW = 1 Hz).</p> Full article ">Figure 16
<p>Comparison of the frequency spectra of the four channels CH1–CH4 and combined signal, one section of the spectra is shown smoothed for better visualization; (RBW = 1 Hz).</p> Full article ">Figure 17
<p>Velocity signal of the four channels CH1–CH4 and the combined signal with poor alignment of the measurement heads; (RBW = 2.5 Hz).</p> Full article ">Figure 18
<p>Frequency spectra of the four channels CH1–CH4 and the combined signal; poor alignment of the measurement heads; (RBW = 2.5 Hz).</p> Full article ">
19 pages, 3912 KiB  
Article
A FOD Detection Approach on Millimeter-Wave Radar Sensors Based on Optimal VMD and SVDD
by Jun Zhong, Xin Gou, Qin Shu, Xing Liu and Qi Zeng
Sensors 2021, 21(3), 997; https://doi.org/10.3390/s21030997 - 2 Feb 2021
Cited by 10 | Viewed by 3524
Abstract
Foreign object debris (FOD) on airport runways can cause serious accidents and huge economic losses. FOD detection systems based on millimeter-wave (MMW) radar sensors have the advantages of higher range resolution and lower power consumption. However, it is difficult for traditional FOD detection [...] Read more.
Foreign object debris (FOD) on airport runways can cause serious accidents and huge economic losses. FOD detection systems based on millimeter-wave (MMW) radar sensors have the advantages of higher range resolution and lower power consumption. However, it is difficult for traditional FOD detection methods to detect and distinguish weak signals of targets from strong ground clutter. To solve this problem, this paper proposes a new FOD detection approach based on optimized variational mode decomposition (VMD) and support vector data description (SVDD). This approach utilizes SVDD as a classifier to distinguish FOD signals from clutter signals. More importantly, the VMD optimized by whale optimization algorithm (WOA) is used to improve the accuracy and stability of the classifier. The results from both the simulation and field case show the excellent FOD detection performance of the proposed VMD-SVDD method. Full article
(This article belongs to the Section Remote Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Time-frequency diagram of LFMCW radar.</p> Full article ">Figure 2
<p>Flowchart of the proposed method.</p> Full article ">Figure 3
<p>Simulation data: (<b>a</b>) Target signals without clutter; (<b>b</b>) Target signals with clutter.</p> Full article ">Figure 4
<p>Optimization process of the WOA.</p> Full article ">Figure 5
<p>Signal decomposition using the optimal VMD.</p> Full article ">Figure 6
<p>IMF selection with the cross-correlation threshold.</p> Full article ">Figure 7
<p>VDM processing with different parameters: (<b>a</b>) VMD parameters are <math display="inline"><semantics> <mrow> <mi>α</mi> <mo>=</mo> <mn>2000</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>K</mi> <mo>=</mo> <mn>7</mn> </mrow> </semantics></math>; (<b>b</b>) VMD parameters are <math display="inline"><semantics> <mrow> <mi>α</mi> <mo>=</mo> <mn>2000</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>K</mi> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math>; (<b>c</b>) VMD parameters are <math display="inline"><semantics> <mrow> <mi>α</mi> <mo>=</mo> <mn>1000</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>K</mi> <mo>=</mo> <mn>7</mn> </mrow> </semantics></math>.</p> Full article ">Figure 8
<p>Mode selection results in four cases: (<b>a</b>) Single target with <math display="inline"><semantics> <mrow> <mi>S</mi> <mi>C</mi> <mi>R</mi> <mo>=</mo> <mo>−</mo> <mrow> <mn>15</mn> <mtext> </mtext> <mi>dB</mi> </mrow> </mrow> </semantics></math>; (<b>b</b>) Two targets with <math display="inline"><semantics> <mrow> <mi>S</mi> <mi>C</mi> <mi>R</mi> <mo>=</mo> <mo>−</mo> <mrow> <mn>15</mn> <mtext> </mtext> <mi>dB</mi> </mrow> </mrow> </semantics></math>; (<b>c</b>) Single target with <math display="inline"><semantics> <mrow> <mi>S</mi> <mi>C</mi> <mi>R</mi> <mo>=</mo> <mo>−</mo> <mrow> <mn>20</mn> <mtext> </mtext> <mi>dB</mi> </mrow> </mrow> </semantics></math>; (<b>d</b>) Two targets with <math display="inline"><semantics> <mrow> <mi>S</mi> <mi>C</mi> <mi>R</mi> <mo>=</mo> <mo>−</mo> <mrow> <mn>20</mn> <mtext> </mtext> <mi>dB</mi> </mrow> </mrow> </semantics></math>.</p> Full article ">Figure 9
<p>Classification results: (<b>a</b>) Training result without VMD processing; (<b>b</b>) Testing result without VMD processing; (<b>c</b>) Training result with VMD processing; (<b>d</b>) Testing result with VMD processing.</p> Full article ">Figure 10
<p>Classification results of clutter data set: (<b>a</b>) Training result; (<b>b</b>) Testing result.</p> Full article ">Figure 11
<p>Detection probabilities of four detection methods.</p> Full article ">Figure 12
<p>Field measure scenario.</p> Full article ">Figure 13
<p>Signals in scene 1: (<b>a</b>) Original signal; (<b>b</b>) Target detection result.</p> Full article ">Figure 14
<p>Signal processing for scene 1: (<b>a</b>) Mode components after VMD processing; (<b>b</b>) SVDD classification result.</p> Full article ">Figure 15
<p>Signals in scene 2: (<b>a</b>) Original signal; (<b>b</b>) Target detection result.</p> Full article ">
18 pages, 2271 KiB  
Article
Comparison of Femtosecond Laser-Assisted and Ultrasound-Assisted Cataract Surgery with Focus on Endothelial Analysis
by Anna Schroeter, Martina Kropp, Zeljka Cvejic, Gabriele Thumann and Bojan Pajic
Sensors 2021, 21(3), 996; https://doi.org/10.3390/s21030996 - 2 Feb 2021
Cited by 17 | Viewed by 3743
Abstract
Femtosecond laser-assisted cataract surgery has the potential to make critical steps of cataract surgery easier and safer, and reduce endothelial cell loss, thus, improving postoperative outcomes. This study compared FLACS with the conventional method in terms of endothelial cells behavior, clinical outcomes, and [...] Read more.
Femtosecond laser-assisted cataract surgery has the potential to make critical steps of cataract surgery easier and safer, and reduce endothelial cell loss, thus, improving postoperative outcomes. This study compared FLACS with the conventional method in terms of endothelial cells behavior, clinical outcomes, and capsulotomy precision. Methods: In a single-center, randomized controlled study, 130 patients with cataracta senilis received FLACS or conventional cataract surgery. Results: A significant endothelial cell loss was observed postoperatively, compared to the preoperative values in both groups. The endothelial cell counts was significantly better in the FLACS group in cataract grade 2 (p = 0.048) patients, compared to conventionally at 4 weeks. The effective phaco time was notably shorter in grade 2 of the FLACS group (p = 0.007) compared to the conventional. However, no statistically significant differences were found for the whole sample, including all cataract grades, due to the overall cataract density in the FLACS group being significantly higher (2.60 ± 0.58, p < 0.001) as compared to conventional methods (2.23 ± 0.42). Conclusions: Low energy FLACS provides a better result compared to endothelial cell loss, size, and shape variations, as well as in effective phaco time within certain cataract grade subgroups. A complete comparison between two groups was not possible because of the higher cataract grade in the FLACS. FLACS displayed a positive effect on endothelial cell preservation and was proven to be much more precise. Full article
(This article belongs to the Special Issue Emerging Technologies for Sensing, Imaging and Biomechanics of the Eye for Diagnostics and Surgery)
►▼ Show Figures

Figure 1

Figure 1
<p>Endothelial cell analysis.</p> Full article ">Figure 2
<p>Intraoperative analyses of all ocular structure.</p> Full article ">Figure 3
<p>The planned cutting lines and areas are marked grey, the red bars are the calculated safety distances.</p> Full article ">Figure 4
<p>Status after femtosecond laser application with eight pie pieces lens fragmentation.</p> Full article ">Figure 5
<p>Distribution of cataract grades in the two groups.</p> Full article ">Figure 6
<p>Best corrected visual acuity (BCVA) over time for both groups.</p> Full article ">
18 pages, 3429 KiB  
Review
Carbon Nanotube Field-Effect Transistor-Based Chemical and Biological Sensors
by Xuesong Yao, Yalei Zhang, Wanlin Jin, Youfan Hu and Yue Cui
Sensors 2021, 21(3), 995; https://doi.org/10.3390/s21030995 - 2 Feb 2021
Cited by 53 | Viewed by 9536
Abstract
Chemical and biological sensors have attracted great interest due to their importance in applications of healthcare, food quality monitoring, environmental monitoring, etc. Carbon nanotube (CNT)-based field-effect transistors (FETs) are novel sensing device configurations and are very promising for their potential to drive many [...] Read more.
Chemical and biological sensors have attracted great interest due to their importance in applications of healthcare, food quality monitoring, environmental monitoring, etc. Carbon nanotube (CNT)-based field-effect transistors (FETs) are novel sensing device configurations and are very promising for their potential to drive many technological advancements in this field due to the extraordinary electrical properties of CNTs. This review focuses on the implementation of CNT-based FETs (CNTFETs) in chemical and biological sensors. It begins with the introduction of properties, and surface functionalization of CNTs for sensing. Then, configurations and sensing mechanisms for CNT FETs are introduced. Next, recent progresses of CNTFET-based chemical sensors, and biological sensors are summarized. Finally, we end the review with an overview about the current application status and the remaining challenges for the CNTFET-based chemical and biological sensors. Full article
(This article belongs to the Special Issue State-of-the-Art Biosensors Technology in China 2020–2021)
►▼ Show Figures

Figure 1

Figure 1
<p>Surface functionalization of carbon nanotubes (CNTs). (<b>a</b>) The introduction of the -COOH group to the surface of the CNT [<a href="#B8-sensors-21-00995" class="html-bibr">8</a>]. (<b>b</b>) The introduction of-COOH on the surface of CNTs by the oxidation reaction [<a href="#B9-sensors-21-00995" class="html-bibr">9</a>]. (<b>c</b>) A metal iridium complex catalyst was coated on the surface of CNTs through non-covalent bond accumulation [<a href="#B10-sensors-21-00995" class="html-bibr">10</a>]. (<b>d</b>) TEM image which shows the nominal chemical structure of the polymer backbone [<a href="#B11-sensors-21-00995" class="html-bibr">11</a>]. (<b>e</b>) The atomic force microscopy image of the waxy corn amylopectin-single-walled CNT (SWCNT) film [<a href="#B12-sensors-21-00995" class="html-bibr">12</a>]. (<b>f</b>) Schematic representation of surfactants adsorb onto the CNT surfaces [<a href="#B13-sensors-21-00995" class="html-bibr">13</a>].</p> Full article ">Figure 2
<p>Illustration of CNT-based field-effect transistors (FETs) for sensing. (<b>a</b>) Schematic representation of a general CNTFET [<a href="#B17-sensors-21-00995" class="html-bibr">17</a>]. (<b>b</b>) Schematic diagram of an electrolyte-gated CNTFET [<a href="#B20-sensors-21-00995" class="html-bibr">20</a>].</p> Full article ">Figure 3
<p>CNTFET-based gas sensor. (<b>a</b>) The response of CNTFET to methanol with (open symbols) and without (filled symbols) KMnO<sub>4</sub> treatment [<a href="#B48-sensors-21-00995" class="html-bibr">48</a>]. (<b>b</b>) The response of polyaniline (PANI) (top) and multi-walled CNT (MWCNT)-PANI (bottom) membranes to methane gas [<a href="#B49-sensors-21-00995" class="html-bibr">49</a>]. (<b>c</b>) The response of SWCNTs with PEDOT:PSS coating to methanol [<a href="#B42-sensors-21-00995" class="html-bibr">42</a>]. (<b>d</b>) Selective response of CNTFET sensor functionalized with Au nanoparticles [<a href="#B52-sensors-21-00995" class="html-bibr">52</a>]. (<b>e</b>) CNTFET ethanol gas sensors functionalized with 10 nm Fe layer [<a href="#B44-sensors-21-00995" class="html-bibr">44</a>]. (<b>f</b>) Aligned CNx nanotubes sensors in response to different vapors [<a href="#B46-sensors-21-00995" class="html-bibr">46</a>].</p> Full article ">Figure 4
<p>CNTFET-based H<sub>2</sub>O<sub>2</sub> sensor. (<b>a</b>) Schematic of a CNTFET for glucose sensing via the detection of H<sub>2</sub>O<sub>2</sub> and (<b>b</b>) its relative resistance change in response to different glucose concentrations [<a href="#B54-sensors-21-00995" class="html-bibr">54</a>]. The change of current of CNTFETs in response to (<b>c</b>) different lactate and (<b>d</b>) glucose concentrations via enzymatic reactions with H<sub>2</sub>O<sub>2</sub> as a by-product [<a href="#B55-sensors-21-00995" class="html-bibr">55</a>].</p> Full article ">Figure 5
<p>CNTFET-based sensors for protein detection. (<b>a</b>) Schematic illustration of a SWCNT device for cytochrome c detection [<a href="#B61-sensors-21-00995" class="html-bibr">61</a>]. (<b>b</b>) Conductance of the CNTFET as a function of the electrochemical potential in phosphate buffer and 200 µM cytochrome c in buffer. (<b>c</b>) Schematic device diagram and (<b>d</b>) current change of CNT nanosensors for prostate-specific antigen (PSA) detection [<a href="#B62-sensors-21-00995" class="html-bibr">62</a>]. (<b>e</b>) Schematic illustration of chromogranin (CgA) released from neurons and (<b>f</b>) its detection by monitoring the current change of a CgA antibody-modified CNTFET [<a href="#B63-sensors-21-00995" class="html-bibr">63</a>].</p> Full article ">Figure 6
<p>CNTFET-based sensors for cell detection. (<b>a</b>) From top to bottom, transfer curves before and after exposure to Salmonella with concentrations of 100 cfu/mL, 300 cfu/mL and 500 cfu/mL, respectively [<a href="#B73-sensors-21-00995" class="html-bibr">73</a>]. (<b>b</b>) Current change of the CNTFET for Escherichia coli detection [<a href="#B74-sensors-21-00995" class="html-bibr">74</a>]. (<b>c</b>) Prostate-specific antigen PSA-ACT complex detection on the CNTFETs modified with a 1:3 linker to spacer ratio [<a href="#B75-sensors-21-00995" class="html-bibr">75</a>].</p> Full article ">Figure 7
<p>CNTFET-based sensors for DNA detection. (<b>a</b>) Working principle of CNTFET-based DNA sensors [<a href="#B91-sensors-21-00995" class="html-bibr">91</a>]. (<b>b</b>) Normalized conductance as a function of target DNA concentrations [<a href="#B87-sensors-21-00995" class="html-bibr">87</a>]. (<b>c</b>) Schematic illustration of DNA detection with reporter DNA conjugated with Au nanoparticles [<a href="#B88-sensors-21-00995" class="html-bibr">88</a>].</p> Full article ">Figure 8
<p>CNTFET-based sensors for virus detection. (<b>a</b>) The experimental setup and (<b>b</b>) transconductance change vs. RNA concentrations of the hepatitis c virus (HCV) sensor [<a href="#B92-sensors-21-00995" class="html-bibr">92</a>]. (<b>c</b>) Detection of H5N1 at different concentrations [<a href="#B93-sensors-21-00995" class="html-bibr">93</a>]. (<b>d</b>) Response to influenza type A virus DNA within 1 min [<a href="#B91-sensors-21-00995" class="html-bibr">91</a>].</p> Full article ">
16 pages, 39882 KiB  
Article
No Reference, Opinion Unaware Image Quality Assessment by Anomaly Detection
by Marco Leonardi, Paolo Napoletano, Raimondo Schettini and Alessandro Rozza
Sensors 2021, 21(3), 994; https://doi.org/10.3390/s21030994 - 2 Feb 2021
Cited by 7 | Viewed by 3377
Abstract
We propose an anomaly detection based image quality assessment method which exploits the correlations between feature maps from a pre-trained Convolutional Neural Network (CNN). The proposed method encodes the intra-layer correlation through the Gram matrix and then estimates the quality score combining the [...] Read more.
We propose an anomaly detection based image quality assessment method which exploits the correlations between feature maps from a pre-trained Convolutional Neural Network (CNN). The proposed method encodes the intra-layer correlation through the Gram matrix and then estimates the quality score combining the average of the correlation and the output from an anomaly detection method. The latter evaluates the degree of abnormality of an image by computing a correlation similarity with respect to a dictionary of pristine images. The effectiveness of the method is tested on different benchmarking datasets (LIVE-itW, KONIQ, and SPAQ). Full article
(This article belongs to the Section Sensing and Imaging)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic view of the proposed method. The intra-layer correlation is computed by the Gram matrix over the activation volumes of a Convolutional Neural Network (CNN). Then the <span class="html-italic">abnormality</span> and the average of the correlation are computed before applying the min-max scaling on both of them. In the end, the two metrics are summed resulting in the predicted image quality score.</p> Full article ">Figure 2
<p>Schematic view of Gram matrix computation. In (<b>a</b>) is reported the feature maps of the <span class="html-italic">j</span>-th layer. (<b>b</b>) illustrate, for some of the indices of the Gram matrix, how they are computed. With the symbol ∘ we refer to the element-wise matrix product.</p> Full article ">Figure 3
<p>Visual overview of the Gram matrix efficient computation and feature vector extrapolation. In (<b>a</b>) is shown how the activation volume is reshaped to efficiently compute the Gram matrix in (<b>b</b>) to finally compute the feature vector that represents the intra-layer correlation.</p> Full article ">Figure 4
<p>Overview of the creation of the dictionary for the degree of abnormality computation. The activation volumes of a given CNN are extracted and then the Gram matrix is computed to get the intra-layer correlation for a subset of pristine images. Subsequently dimensionality reduction is applied through Principal Component Analysis (PCA). Finally Mean Shift algorithm is performed to compute clusters on which centroids are then extracted as entry of the dictionary.</p> Full article ">Figure 5
<p>Estimated density distribution of Mean Opinion Scores for the three datasets: (<b>a</b>) LIVE in the Wild Image Quality Challenge Database (LIVE-itW), (<b>b</b>) KonIQ-10k (KONIQ), and (<b>c</b>) Smartphone Photography Attribute and Quality database (SPAQ). The bars represents the normalized histogram, the blue line is the estimated density distribution while in red line is the 75th percentile respect the Mean Opinion Score (MOS).</p> Full article ">Figure 6
<p>Sample images form the three Image Quality Assessment (IQA) databases: (<b>a</b>) and (<b>d</b>) images from the LIVE-itW with a MOS of 78.81 and 43.67 respectively, (<b>b</b>) and (<b>e</b>) KONIQ’s pictures with a MOS of 71.46 and 43.36; finally (<b>c</b>) and (<b>f</b>) photos form SPAQ whit a MOS of 75.43 and 33.0 respectively.</p> Full article ">Figure 7
<p>Sample images form the KADIS700k databases. (<b>a</b>–<b>c</b>) random photos from KADIS700k.</p> Full article ">Figure 8
<p>Scatter plots of the predicted quality scores respect MOS for the three considered datasets: (<b>a</b>) LIVE-itW, (<b>b</b>) KONIQ, and (<b>c</b>) SPAQ. In red is depicted the second-order interpolation line. The points in blue belongs to the bad quality images (MOS &lt; 75° percentile of the MOS distribution for that dataset) while the orange ones refer to the good quality images.</p> Full article ">Figure 9
<p>Examples of predicted quality score (QS) alongside the mean opinion score (MOS). First column images (<b>a</b>,<b>d</b>) belong to LIVE-itW database, second column picures (<b>b</b>,<b>e</b>) are from KONIQ dataset while the last column photos (<b>c</b>,<b>f</b>) belong to the SPAQ collection.</p> Full article ">
17 pages, 2544 KiB  
Article
Structural Health Monitoring Using Ultrasonic Guided-Waves and the Degree of Health Index
by Sergio Cantero-Chinchilla, Gerardo Aranguren, José Manuel Royo, Manuel Chiachío, Josu Etxaniz and Andrea Calvo-Echenique
Sensors 2021, 21(3), 993; https://doi.org/10.3390/s21030993 - 2 Feb 2021
Cited by 26 | Viewed by 5245
Abstract
This paper proposes a new damage index named degree of health (DoH) to efficiently tackle structural damage monitoring in real-time. As a key contribution, the proposed index relies on a pattern matching methodology that measures the time-of-flight mismatch of sequential ultrasonic guided-wave measurements [...] Read more.
This paper proposes a new damage index named degree of health (DoH) to efficiently tackle structural damage monitoring in real-time. As a key contribution, the proposed index relies on a pattern matching methodology that measures the time-of-flight mismatch of sequential ultrasonic guided-wave measurements using fuzzy logic fundamentals. The ultrasonic signals are generated using the transmission beamforming technique with a phased-array of piezoelectric transducers. The acquisition is carried out by two phased-arrays to compare the influence of pulse-echo and pitch-catch modes in the damage assessment. The proposed monitoring approach is illustrated in a fatigue test of an aluminum sheet with an initial notch. As an additional novelty, the proposed pattern matching methodology uses the data stemming from the transmission beamforming technique for structural health monitoring. The results demonstrate the efficiency and robustness of the proposed framework in providing a qualitative and quantitative assessment for fatigue crack damage. Full article
(This article belongs to the Special Issue Structural Health Monitoring with Ultrasonic Guided-Waves Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Panel (<b>a</b>): Selection of characteristic points (CPs) above the threshold value <math display="inline"><semantics> <msub> <mi>A</mi> <mi>t</mi> </msub> </semantics></math>. Panel (<b>b</b>): Illustration of ToF dispersion due to repeated measurements. Panel (<b>c</b>): Trapezoidal function used to evaluate the ToF mismatch based on the repeated measurements of one CP (blue circles).</p> Full article ">Figure 2
<p>Schematic workflow of the methodology divided between the data acquisition of ultrasonic data and its post-processing depending on the actual structural state (i.e., pristine or in operation).</p> Full article ">Figure 3
<p>Panel (<b>a</b>): Schematic of the specimen and notch geometry, along with the position of the piezoelectric wafer active sensors (PWAS) arrays. Panel (<b>b</b>): Picture of M(T) aluminum specimen with two permanently attached phased-arrays mounted on the fatigue testing machine.</p> Full article ">Figure 4
<p>Schematic of the ultrasonic guided-wave based tests.</p> Full article ">Figure 5
<p>Time (panel (<b>a</b>)) and frequency (panel (<b>b</b>)) domain representations of the bandpass filter.</p> Full article ">Figure 6
<p>Degree of health (DoH) matrices at different fatigue cycles for both phased-arrays, that is, pulse-echo (T<math display="inline"><semantics> <msub> <mrow/> <mi>i</mi> </msub> </semantics></math>) and pitch-catch (S<math display="inline"><semantics> <msub> <mrow/> <mi>i</mi> </msub> </semantics></math>).</p> Full article ">Figure 7
<p>Evolution of the mean value of all DoH matrices obtained throughout the fatigue test in comparison with the crack length of the aluminum specimen in panel (<b>a</b>). Panel (<b>b</b>) shows the fatigue crack at 100,000 cycles along with the points used to digitize its length.</p> Full article ">Figure 8
<p>Evolution of M<math display="inline"><semantics> <msup> <mrow/> <mi>j</mi> </msup> </semantics></math> for the PWAS of the arrays in pulse-echo (panels (<b>a</b>,<b>b</b>)) and pitch-catch (panels (<b>c</b>,<b>d</b>)) modes at two particular directions.</p> Full article ">Figure 9
<p>Relationship between crack growth and mean of DoH for both pulse-echo and pitch-catch working modes. Panels (<b>a</b>,<b>b</b>) provide the data for the 37 directions with an average value of DoH with respect to the PWAS in the array. Additionally, (<b>c</b>,<b>d</b>) show the mean and 90% and 50% uncertainty bands of the DoH data.</p> Full article ">Figure 10
<p>Crack growth data for QQ-A-250/5 ‘O’ M(T) aluminum specimen correlated with the stress intensity range.</p> Full article ">
18 pages, 3865 KiB  
Article
The Application of Electromagnetic Sensors for Determination of Cherenkov Cone Inside and in the Vicinity of the Detector Volume in Any Environment Known
by Valeriu Savu, Mădălin Ion Rusu and Dan Savastru
Sensors 2021, 21(3), 992; https://doi.org/10.3390/s21030992 - 2 Feb 2021
Viewed by 2361
Abstract
The neutrinos of cosmic radiation, due to interaction with any known medium in which the Cherenkov detector is used, produce energy radiation phenomena in the form of a Cherenkov cone, in very large frequency spectrum. These neutrinos carry with them the information about [...] Read more.
The neutrinos of cosmic radiation, due to interaction with any known medium in which the Cherenkov detector is used, produce energy radiation phenomena in the form of a Cherenkov cone, in very large frequency spectrum. These neutrinos carry with them the information about the phenomena that produced them and by detecting the electromagnetic energies generated by the Cherenkov cone, we can find information about the phenomena that formed in the universe, at a much greater distance, than possibility of actually detection with current technologies. At present, a very high number of sensors for detection electromagnetic energy is required. Thus, some sensors may detect very low energy levels, which can lead to the erroneous determination of the Cherenkov cone, thus leading to information errors. As a novelty, we propose, to use these sensors for determination of the dielectrically permittivity of any known medium in which the Cherenkov detector is used, by preliminary measurements, the subsequent simulation of the data and the reconstruction of the Cherenkov cone, leading to a significant reduction of problems and minimizing the number of sensors, implicitly the cost reductions. At the same time, we offer the possibility of reconstructing the Cherenkov cone outside the detector volume. Full article
(This article belongs to the Special Issue State-of-the-Art Sensors Technology in Romania 2020)
►▼ Show Figures

Figure 1

Figure 1
<p>Neutrinos detector in a salt block consisting of electromagnetic sensors placed in a cube with a 1 km side [<a href="#B23-sensors-21-00992" class="html-bibr">23</a>].</p> Full article ">Figure 2
<p>The measurement system inserted into the salt block (horizontal or vertical plane).</p> Full article ">Figure 3
<p>The dimensions of the emission element.</p> Full article ">Figure 4
<p>Electrical diagram of the BAL-UN and the symmetrizer (* represents the beginning of the inductor winding).</p> Full article ">Figure 5
<p>Electrical diagram of the adapter “in <span class="html-italic">Π</span>”.</p> Full article ">Figure 6
<p>Practical mounting diagram for adjusting the impedance adapter “in <span class="html-italic">Π</span>”.</p> Full article ">Figure 7
<p>Air measurements.</p> Full article ">Figure 8
<p>The emission element introduced in saline environment for vertical plane.</p> Full article ">Figure 9
<p>Reception electromagnetic sensor introduced in saline environment for vertical plane.</p> Full article ">Figure 10
<p>The measurement result for 187.5 MHz frequency for vertical plane.</p> Full article ">Figure 11
<p>Diagrams associated with the after-effect of the electric field applied to linear dielectrics at: (<b>a</b>) sudden variations, (<b>b</b>) sinusoidal [<a href="#B42-sensors-21-00992" class="html-bibr">42</a>].</p> Full article ">
23 pages, 5759 KiB  
Article
Social Collective Attack Model and Procedures for Large-Scale Cyber-Physical Systems
by Peidong Zhu, Peng Xun, Yifan Hu and Yinqiao Xiong
Sensors 2021, 21(3), 991; https://doi.org/10.3390/s21030991 - 2 Feb 2021
Cited by 1 | Viewed by 2991
Abstract
A large-scale Cyber-Physical System (CPS) such as a smart grid usually provides service to a vast number of users as a public utility. Security is one of the most vital aspects in such critical infrastructures. The existing CPS security usually considers the attack [...] Read more.
A large-scale Cyber-Physical System (CPS) such as a smart grid usually provides service to a vast number of users as a public utility. Security is one of the most vital aspects in such critical infrastructures. The existing CPS security usually considers the attack from the information domain to the physical domain, such as injecting false data to damage sensing. Social Collective Attack on CPS (SCAC) is proposed as a new kind of attack that intrudes into the social domain and manipulates the collective behavior of social users to disrupt the physical subsystem. To provide a systematic description framework for such threats, we extend MITRE ATT&CK, the most used cyber adversary behavior modeling framework, to cover social, cyber, and physical domains. We discuss how the disinformation may be constructed and eventually leads to physical system malfunction through the social-cyber-physical interfaces, and we analyze how the adversaries launch disinformation attacks to better manipulate collective behavior. Finally, simulation analysis of SCAC in a smart grid is provided to demonstrate the possibility of such an attack. Full article
(This article belongs to the Section Sensor Networks)
►▼ Show Figures

Figure 1

Figure 1
<p>Demand response model.</p> Full article ">Figure 2
<p>Botnet and system instability: attack from the cyber domain.</p> Full article ">Figure 3
<p>Price modification and system instability.</p> Full article ">Figure 4
<p>Social-cyber-physical: attack initiated from the social domain.</p> Full article ">Figure 5
<p>Cyber-Physical System (CPS) model with social domains.</p> Full article ">Figure 6
<p>Steps of Social Collective Attack on CPS (SCAC).</p> Full article ">Figure 7
<p>User behavior model in SCAC.</p> Full article ">Figure 8
<p>An example of a fast attack.</p> Full article ">Figure 9
<p>An example of the gradual reverse demands attack.</p> Full article ">Figure 10
<p>Process of the stability control of the physical system.</p> Full article ">Figure 11
<p>State transition model [<a href="#B33-sensors-21-00991" class="html-bibr">33</a>].</p> Full article ">Figure 12
<p>A simplified model of the power grid system.</p> Full article ">Figure 13
<p>Comparing the effects of the disinformation attack with <math display="inline"><semantics> <mrow> <mi>α</mi> <mo>=</mo> <mn>0.8</mn> </mrow> </semantics></math> and with <math display="inline"><semantics> <mrow> <mi>α</mi> <mo>=</mo> <mn>0.2</mn> </mrow> </semantics></math>, where <math display="inline"><semantics> <mi>β</mi> </semantics></math> = 0.5 and initial demand is 2400 MW.</p> Full article ">Figure 14
<p>Attack effects of real changing demands and evaluated changing demands with the change of parameter <math display="inline"><semantics> <mi>α</mi> </semantics></math>.</p> Full article ">Figure 15
<p>The relationship among initial demands, changed demands, and the attack effect.</p> Full article ">Figure 16
<p>Comparing the attack effects between Attack Type I and Attack Type II, where <math display="inline"><semantics> <mi>α</mi> </semantics></math> = 0.5 and initial demands = 2400 MW.</p> Full article ">
20 pages, 4036 KiB  
Article
On the Use of Movement-Based Interaction with Smart Textiles for Emotion Regulation
by Mengqi Jiang, Vijayakumar Nanjappan, Martijn ten Bhömer and Hai-Ning Liang
Sensors 2021, 21(3), 990; https://doi.org/10.3390/s21030990 - 2 Feb 2021
Cited by 6 | Viewed by 5665
Abstract
Research from psychology has suggested that body movement may directly activate emotional experiences. Movement-based emotion regulation is the most readily available but often underutilized strategy for emotion regulation. This research aims to investigate the emotional effects of movement-based interaction and its sensory feedback [...] Read more.
Research from psychology has suggested that body movement may directly activate emotional experiences. Movement-based emotion regulation is the most readily available but often underutilized strategy for emotion regulation. This research aims to investigate the emotional effects of movement-based interaction and its sensory feedback mechanisms. To this end, we developed a smart clothing prototype, E-motionWear, which reacts to four movements (elbow flexion/extension, shoulder flexion/extension, open and closed arms, neck flexion/extension), fabric-based detection sensors, and three-movement feedback mechanisms (audio, visual and vibrotactile). An experiment was conducted using a combined qualitative and quantitative approach to collect participants’ objective and subjective emotional feelings. Results indicate that there was no interaction effect between movement and feedback mechanism on the final emotional results. Participants preferred vibrotactile and audio feedback rather than visual feedback when performing these four kinds of upper body movements. Shoulder flexion/extension and open-closed arm movements were more effective for improving positive emotion than elbow flexion/extension movements. Participants thought that the E-motionWear prototype were comfortable to wear and brought them new emotional experiences. From these results, a set of guidelines were derived that can help frame the design and use of smart clothing to support users’ emotional regulation. Full article
(This article belongs to the Special Issue Wearable Sensors for Healthcare)
►▼ Show Figures

Figure 1

Figure 1
<p>The body movements detected using our prototype: (<b>a</b>) neck flexion/extension; (<b>b</b>) elbow flexion/extension; (<b>c</b>) shoulder flexion/extension; and (<b>d</b>) open and closed arms. Green dots indicate the sensors’ position.</p> Full article ">Figure 2
<p>Two types of fabric sensors used in our prototype: (<b>a</b>) Neck flexion/extension fabric sensor, made of two kinds of conductive fabrics; (<b>b</b>) Arm, elbow, and shoulder movements’ fabric sensor, made of conductive knitted elastic fabric.</p> Full article ">Figure 3
<p>(<b>a</b>) Illustrative image of our E-motionWear t-shirt with details of the components; (<b>b</b>) Zigzag sewing to protect the conductive thread circuit; and (<b>c</b>) Stainless conductive thread employed in our prototype.</p> Full article ">Figure 4
<p>The female version of the E-MotionWear prototype and placement of the different sensors.</p> Full article ">Figure 5
<p>Illustration of the experiment apparatus and set up.</p> Full article ">Figure 6
<p>SAM tests that were used in the experiment process.</p> Full article ">Figure 7
<p>The normalized confusion matrix results for movement recognition of the E-motionWear prototype: (<b>a</b>) visual, (<b>b</b>) vibrotactile, and (<b>c</b>) audio feedback mechanisms.</p> Full article ">Figure 8
<p>Estimated Marginal Means of SAM questionnaire data when participants perform under different movements and feedback mechanisms. (<b>a</b>) SAM valence—the intrinsic positive or negative valence of their emotional feelings; and (<b>b</b>) SAM arousal—the emotional state of being awakened.</p> Full article ">Figure 9
<p>An example of the output of the AffdexMe App on a facial expression during the experiment: (<b>a</b>) initial position and (<b>b</b>) while performing the open arm movement.</p> Full article ">Figure 10
<p>Estimated Marginal Means of the AffdexMe facial data when participants perform under different movements and feedback mechanisms. (<b>a</b>) Joy—the frequency of happy emotion during the movement performance; (<b>b</b>) Positive valence—the frequency of positive emotion; (<b>c</b>) Surprise—the frequency of surprise emotions; and (<b>d</b>) Engagement—the frequency of emotional engagement.</p> Full article ">Figure 11
<p>The mean values of the six elements of comfort rating scales for E-motionWear.</p> Full article ">Figure A1
<p>The processed signals of the four movements from the textile sensors.</p> Full article ">
16 pages, 4026 KiB  
Article
3D Multiple-Antenna Channel Modeling and Propagation Characteristics Analysis for Mobile Internet of Things
by Wenbo Zeng, Yigang He, Bing Li and Shudong Wang
Sensors 2021, 21(3), 989; https://doi.org/10.3390/s21030989 - 2 Feb 2021
Cited by 5 | Viewed by 2366
Abstract
The demand for optimization design and performance evaluation of wireless communication links in a mobile Internet of Things (IoT) motivates the exploitation of realistic and tractable channel models. In this paper, we develop a novel three-dimensional (3D) multiple-antenna channel model to adequately characterize [...] Read more.
The demand for optimization design and performance evaluation of wireless communication links in a mobile Internet of Things (IoT) motivates the exploitation of realistic and tractable channel models. In this paper, we develop a novel three-dimensional (3D) multiple-antenna channel model to adequately characterize the scattering environment for mobile IoT scenarios. Specifically, taking into consideration both accuracy and mathematical tractability, a 3D double-spheres model and ellipsoid model are introduced to describe the distribution region of the local scatterers and remote scatterers, respectively. Based on the explicit geometry relationships between transmitter, receiver, and scatterers, we derive the complex channel gains by adopting the radio-wave propagation model. Subsequently, the correlation-based approach for theoretical analysis is performed, and the detailed impacts with respect to the antenna deployment, scatterer distribution, and scatterer density on the vital statistical properties are investigated. Numerical simulation results have shown that the statistical channel characteristics in the developed simulation model nicely match those of the corresponding theoretical results, which demonstrates the utility of our model. Full article
(This article belongs to the Section Remote Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Proposed 3D double-spheres model capturing local scatterer distribution for mobile Internet of Things (IoT) scenarios.</p> Full article ">Figure 2
<p>Proposed 3D ellipsoid model capturing remote scatterer distribution for mobile IoT scenarios.</p> Full article ">Figure 3
<p>The illustration of the continuous random variables <math display="inline"><semantics> <msubsup> <mi>α</mi> <mi>T</mi> <mn>1</mn> </msubsup> </semantics></math>, <math display="inline"><semantics> <msubsup> <mi>α</mi> <mi>R</mi> <mn>2</mn> </msubsup> </semantics></math>, <math display="inline"><semantics> <msubsup> <mi>α</mi> <mi>R</mi> <mn>3</mn> </msubsup> </semantics></math>, <math display="inline"><semantics> <msubsup> <mi>β</mi> <mi>T</mi> <mn>1</mn> </msubsup> </semantics></math>, <math display="inline"><semantics> <msubsup> <mi>β</mi> <mi>R</mi> <mn>2</mn> </msubsup> </semantics></math>, and <math display="inline"><semantics> <msubsup> <mi>β</mi> <mi>R</mi> <mn>3</mn> </msubsup> </semantics></math> for the distribution of effective scatterers in the proposed theoretical model.</p> Full article ">Figure 4
<p>Analytical results of the spatial correlation function (CF) <span class="html-italic">ρ<sub>pq</sub></span><sub>,<span class="html-italic">p′q′</span></sub>(<span class="html-italic">δ<sub>T</sub></span>, <span class="html-italic">δ<sub>R</sub></span>) of the proposed model in an isotropic scattering environment for different antenna spacings <span class="html-italic">δ</span>/<span class="html-italic">λ</span> and different antenna array elevation angles <span class="html-italic">ψ</span>.</p> Full article ">Figure 5
<p>Analytical results of the spatial CF <span class="html-italic">ρ<sub>pq</sub></span><sub>,<span class="html-italic">p′q′</span></sub>(<span class="html-italic">δ<sub>T</sub></span>, <span class="html-italic">δ<sub>R</sub></span>) of the proposed model in an isotropic scattering environment for different antenna spacings <span class="html-italic">δ</span>/<span class="html-italic">λ</span> and different maximum ranges of elevation angle deviation <span class="html-italic">β<sub>M</sub></span>.</p> Full article ">Figure 6
<p>Analytical results of the spatial CF <span class="html-italic">ρ<sub>pq</sub></span><sub>,<span class="html-italic">p′q′</span></sub>(<span class="html-italic">δ<sub>T</sub></span>, <span class="html-italic">δ<sub>R</sub></span>) of the proposed model in an isotropic scattering environment for different antenna spacings <span class="html-italic">δ</span>/<span class="html-italic">λ</span> and different antenna array orientation maximum ranges of elevation angle <span class="html-italic">θ</span>.</p> Full article ">Figure 7
<p>Analytical results of the spatial CF <span class="html-italic">ρ<sub>pq</sub></span><sub>,<span class="html-italic">p′q′</span></sub>(<span class="html-italic">δ<sub>T</sub></span>, <span class="html-italic">δ<sub>R</sub></span>) of the proposed model for different antenna spacings <span class="html-italic">δ</span>/<span class="html-italic">λ</span> and different control factors for the distribution concentration <span class="html-italic">k</span>.</p> Full article ">Figure 8
<p>Analytical results of the spatial CF <span class="html-italic">ρ<sub>pq</sub></span><sub>,<span class="html-italic">p′q′</span></sub>(<span class="html-italic">δ<sub>T</sub></span>, <span class="html-italic">δ<sub>R</sub></span>) of the proposed model in a non-isotropic scattering environment for different antenna spacings <span class="html-italic">δ</span>/<span class="html-italic">λ</span> and different mean angles <span class="html-italic">α</span><sub>0</sub> with (<b>a</b>) <span class="html-italic">θ<sub>T</sub></span> = <span class="html-italic">θ<sub>R</sub></span> = <span class="html-italic">0</span> and (<b>b</b>) <span class="html-italic">θ<sub>T</sub></span> = <span class="html-italic">θ<sub>R</sub></span> = <span class="html-italic">π</span>/4.</p> Full article ">Figure 9
<p>Analytical results of the spatial–temporal correlation function (ST-CF) <span class="html-italic">ρ<sub>pq</sub></span><sub>,<span class="html-italic">p′q′</span></sub>(<span class="html-italic">δ<sub>T</sub></span>, <span class="html-italic">δ<sub>R</sub>,</span><span class="html-italic">τ</span>) of the proposed model for low scatterer density and high scatterer density.</p> Full article ">Figure 10
<p>(<b>a</b>) ST-CFs of theoretical model in an isotropic scattering environment, (<b>b</b>) ST-CFs of simulation model in an isotropic scattering environment, and (<b>c</b>) the corresponding absolute error between (<b>a</b>,<b>b</b>).</p> Full article ">Figure 11
<p>(<b>a</b>) ST-CFs of theoretical model in a non-isotropic scattering environment, (<b>b</b>) ST-CFs of simulation model in a non-isotropic scattering environment, and (<b>c</b>) the corresponding absolute error between (<b>a</b>,<b>b</b>).</p> Full article ">
37 pages, 17730 KiB  
Article
Activity Recognition in Residential Spaces with Internet of Things Devices and Thermal Imaging
by Kshirasagar Naik, Tejas Pandit, Nitin Naik and Parth Shah
Sensors 2021, 21(3), 988; https://doi.org/10.3390/s21030988 - 2 Feb 2021
Cited by 19 | Viewed by 5056
Abstract
In this paper, we design algorithms for indoor activity recognition and 3D thermal model generation using thermal images, RGB images, captured from external sensors, and the internet of things setup. Indoor activity recognition deals with two sub-problems: Human activity and household activity recognition. [...] Read more.
In this paper, we design algorithms for indoor activity recognition and 3D thermal model generation using thermal images, RGB images, captured from external sensors, and the internet of things setup. Indoor activity recognition deals with two sub-problems: Human activity and household activity recognition. Household activity recognition includes the recognition of electrical appliances and their heat radiation with the help of thermal images. A FLIR ONE PRO camera is used to capture RGB-thermal image pairs for a scene. Duration and pattern of activities are also determined using an iterative algorithm, to explore kitchen safety situations. For more accurate monitoring of hazardous events such as stove gas leakage, a 3D reconstruction approach is proposed to determine the temperature of all points in the 3D space of a scene. The 3D thermal model is obtained using the stereo RGB and thermal images for a particular scene. Accurate results are observed for activity detection, and a significant improvement in the temperature estimation is recorded in the 3D thermal model compared to the 2D thermal image. Results from this research can find applications in home automation, heat automation in smart homes, and energy management in residential spaces. Full article
(This article belongs to the Section Internet of Things)
►▼ Show Figures

Figure 1

Figure 1
<p>Experiment Setup.</p> Full article ">Figure 2
<p>Interactions among the entities.</p> Full article ">Figure 3
<p>Workflow diagram for Image Annotations.</p> Full article ">Figure 4
<p>Flow diagram for Algorithm 1 (Generating Image Annotation).</p> Full article ">Figure 5
<p>Flow diagram for Algorithm 2 (Generating Video Annotation).</p> Full article ">Figure 6
<p>Flow diagram for Algorithm 3 (Heat Activity Recognition).</p> Full article ">Figure 7
<p>Flow diagram for Algorithm 4 (Human Activity Recognition).</p> Full article ">Figure 8
<p>Flow diagram for Algorithm 5 (Activity Duration).</p> Full article ">Figure 9
<p>Flow Diagram for 3D thermal model.</p> Full article ">Figure 10
<p>Flow diagram for Algorithm 6 (3D Mesh Generation).</p> Full article ">Figure 11
<p>Flow diagram for Algorithm 7 (Temperature Extraction).</p> Full article ">Figure 12
<p>(<b>a</b>) Visual image, (<b>b</b>) Thermal image, and (<b>c</b>) Object detection of a stove/oven.</p> Full article ">Figure 13
<p>(<b>a</b>) Visual image, (<b>b</b>) Thermal image, and (<b>c</b>) Object detection in a living room.</p> Full article ">Figure 14
<p>Refrigerator temperature fluctuations.</p> Full article ">Figure 15
<p>(<b>a</b>) Visual image, (<b>b</b>) Thermal image, and (<b>c</b>) 3D thermal model of a cupboard shelf.</p> Full article ">Figure 16
<p>(<b>a</b>) Visual image, (<b>b</b>) Thermal image, and (<b>c</b>) 3D thermal model of a stove.</p> Full article ">
22 pages, 1978 KiB  
Article
Influence of Noise-Limited Censored Path Loss on Model Fitting and Path Loss-Based Positioning
by Aki Karttunen, Mikko Valkama and Jukka Talvitie
Sensors 2021, 21(3), 987; https://doi.org/10.3390/s21030987 - 2 Feb 2021
Viewed by 2293
Abstract
Positioning is considered one of the key features in various novel industry verticals in future radio systems. Since path loss (PL) or received signal strength-based measurements are widely available in the majority of wireless standards, PL-based positioning has an important role among positioning [...] Read more.
Positioning is considered one of the key features in various novel industry verticals in future radio systems. Since path loss (PL) or received signal strength-based measurements are widely available in the majority of wireless standards, PL-based positioning has an important role among positioning technologies. Conventionally, PL-based positioning has two phases—fitting a PL model to training data and positioning based on the link distance estimates. However, in both phases, the maximum measurable PL is limited by measurement noise. Such immeasurable samples are called censored PL data and such noisy data are commonly neglected in both the model fitting and in the positioning phase. In the case of censored PL, the loss is known to be above a known threshold level and that information can be used in model fitting and in the positioning phase. In this paper, we examine and propose how to use censored PL data in PL model-based positioning. Additionally, we demonstrate with several simulations the potential of the proposed approach for considerable improvements in positioning accuracy (23–57%) and improved robustness against PL model fitting errors. Full article
(This article belongs to the Special Issue Selected Papers from 10th International Conference on Localization and GNSS 2020 (ICL-GNSS 2020))
►▼ Show Figures

Figure 1

Figure 1
<p>Likelihood function illustration example in 1D: measured path loss (PL) of 130 dB to BS1, censored PL <math display="inline"><semantics> <mrow> <mo>&gt;</mo> <mn>140</mn> </mrow> </semantics></math> dB to BS2, and the combined likelihood function. Note that one measured and one censored PL are sufficient to get a unique positioning solution and the same is true in 2D and 3D.</p> Full article ">Figure 2
<p>Antenna gain as a function of the offset angle <math display="inline"><semantics> <mi>θ</mi> </semantics></math>. The directive beam patterns have beam-width of <math display="inline"><semantics> <mrow> <msub> <mi>θ</mi> <mrow> <mn>3</mn> <mi>dB</mi> </mrow> </msub> <mo>=</mo> <msup> <mn>360</mn> <mo>∘</mo> </msup> <mo>/</mo> <mi>N</mi> </mrow> </semantics></math>.</p> Full article ">Figure 3
<p>Fitting the path loss model to the training data with OLS (green) and MLE (red)—Example 1 (<math display="inline"><semantics> <mrow> <mi>α</mi> <mo>=</mo> <mn>4</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>β</mi> <mo>=</mo> <mn>60</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>σ</mi> <mo>=</mo> <mn>7</mn> </mrow> </semantics></math>) and Example 2 (<math display="inline"><semantics> <mrow> <mi>α</mi> <mo>=</mo> <mn>2.2</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <mi>β</mi> <mo>=</mo> <mn>51</mn> </mrow> </semantics></math>). Training data with <math display="inline"><semantics> <mrow> <mi>P</mi> <mi>L</mi> </mrow> </semantics></math> under the noise threshold are shown with black dots, solid lines show <math display="inline"><semantics> <mrow> <mover> <mrow> <mi>P</mi> <mi>L</mi> </mrow> <mo>¯</mo> </mover> <mrow> <mo>(</mo> <mi>d</mi> <mo>)</mo> </mrow> </mrow> </semantics></math>, and the dash lines are <math display="inline"><semantics> <mrow> <mover> <mrow> <mi>P</mi> <mi>L</mi> </mrow> <mo>¯</mo> </mover> <mrow> <mo>(</mo> <mi>d</mi> <mo>)</mo> </mrow> <mo>±</mo> <mn>1.96</mn> <mover accent="true"> <mi>σ</mi> <mo>^</mo> </mover> </mrow> </semantics></math>.</p> Full article ">Figure 4
<p>Fitting the path loss model to the training data with OLS (green) and MLE (red); Example 3, 2 GHz LOS, 2 GHz NLOS, 28 GHz LOS, 28 GHz NLOS. Training data with <math display="inline"><semantics> <mrow> <mi>P</mi> <mi>L</mi> </mrow> </semantics></math> under the noise threshold are shown with black dots, solid lines show <math display="inline"><semantics> <mrow> <mover> <mrow> <mi>P</mi> <mi>L</mi> </mrow> <mo>¯</mo> </mover> <mrow> <mo>(</mo> <mi>d</mi> <mo>)</mo> </mrow> </mrow> </semantics></math>, and the dash lines are <math display="inline"><semantics> <mrow> <mover> <mrow> <mi>P</mi> <mi>L</mi> </mrow> <mo>¯</mo> </mover> <mrow> <mo>(</mo> <mi>d</mi> <mo>)</mo> </mrow> <mo>±</mo> <mn>1.96</mn> <mover accent="true"> <mi>σ</mi> <mo>^</mo> </mover> </mrow> </semantics></math>.</p> Full article ">Figure 5
<p>Likelihood function illustrations: (<b>a</b>) measured PL and omnidirectional antenna, (<b>b</b>) censored PL and omnidirectional antenna, (<b>c</b>) measured PL and directive beam (<math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>), (<b>d</b>) censored PL and directive beam (<math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>). White is likely, gray is possible, and black is an unlikely location.</p> Full article ">Figure 6
<p>Likelihood function illustrations: true location (+), position estimate (×), base stations (BSs; stars), contact with three BSs with directive antennas (<math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>) at <math display="inline"><semantics> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mi>BS</mi> </msub> <mo>,</mo> <msub> <mi>y</mi> <mi>BS</mi> </msub> <mo>)</mo> </mrow> </semantics></math> = (302,324), (547,363), and (146,516). White is likely, gray is possible, and black is an unlikely location.(<b>a</b>) OLS-OPT: ordinary least squares fitting and trilateration positioning. (<b>b</b>) MLE-OPT: Tobit MLE fitting and the ordinary trilateration positioning. (<b>c</b>) OLS-MLE: ordinary least squares fitting and the Tobit MLE positioning. (<b>d</b>) MLE-MLE: Tobit MLE fitting and the Tobit MLE positioning.</p> Full article ">Figure 7
<p>CDF of positioning error of Example 1. Model fitting to training data with ordinary least squares (OLS-) or Tobit MLE fitting (MLE-) and positioning with either ordinary trilateration (-OTP) of Tobit MLE positioning (-MLE). (<b>a</b>) <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>1</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>2</mn> </mrow> </semantics></math>; (<b>b</b>) <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>2</mn> </mrow> </semantics></math>; (<b>c</b>) <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math>.</p> Full article ">Figure 8
<p>CDF of positioning error of Example 2. Model fitting to training data with ordinary least squares (OLS-) or Tobit MLE fitting (MLE-) and positioning with either ordinary trilateration (-OTP) of Tobit MLE positioning (-MLE). (<b>a</b>) <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>1</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math>; (<b>b</b>) <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math>; (<b>c</b>) <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>10</mn> </mrow> </semantics></math>.</p> Full article ">Figure 9
<p>CDF of positioning error of Example 3. Model fitting to training data with ordinary least squares (OLS-) or Tobit MLE fitting (MLE-) and positioning with either ordinary trilateration (-OTP) of Tobit MLE positioning (-MLE). (<b>a</b>) 2 GHz, <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>1</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math>, (<b>b</b>) 2 GHz, <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math>, (<b>c</b>) 28 GHz, <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>5</mn> </mrow> </semantics></math>, (<b>d</b>) 28 GHz, <math display="inline"><semantics> <mrow> <mi>N</mi> <mo>=</mo> <mn>8</mn> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mover> <mi>N</mi> <mo>¯</mo> </mover> <mi>BS</mi> </msub> <mo>=</mo> <mn>10</mn> </mrow> </semantics></math>.</p> Full article ">
17 pages, 8244 KiB  
Article
Use of Multiple Bacteriophage-Based Structural Color Sensors to Improve Accuracy for Discrimination of Geographical Origins of Agricultural Products
by Daun Seol, Daeil Jang, Kyungjoon Cha, Jin-Woo Oh and Hoeil Chung
Sensors 2021, 21(3), 986; https://doi.org/10.3390/s21030986 - 2 Feb 2021
Cited by 2 | Viewed by 2525
Abstract
A single M13 bacteriophage color sensor was previously utilized for discriminating the geographical origins of agricultural products (garlic, onion, and perilla). The resulting discrimination accuracy was acceptable, ranging from 88.6% to 94.0%. To improve the accuracy further, the use of three separate M13 [...] Read more.
A single M13 bacteriophage color sensor was previously utilized for discriminating the geographical origins of agricultural products (garlic, onion, and perilla). The resulting discrimination accuracy was acceptable, ranging from 88.6% to 94.0%. To improve the accuracy further, the use of three separate M13 bacteriophage color sensors containing different amino acid residues providing unique individual color changes (Wild sensor: glutamic acid (E)-glycine (G)-aspartic acid (D), WHW sensor: tryptophan (W)-histidine (H)-tryptophan (W), 4E sensor: four repeating glutamic acids (E)) was proposed. This study was driven by the possibility of enhancing sample discrimination by combining mutually characteristic and complimentary RGB signals obtained from each color sensor, which resulted from dissimilar interactions of sample odors with the employed color sensors. When each color sensor was used individually, the discrimination accuracy based on support vector machine (SVM) ranged from 91.8–94.0%, 88.6–90.3%, and 89.8–92.1% for garlic, onion, and perilla samples, respectively. Accuracy improved to 98.0%, 97.5%, and 97.1%, respectively, by integrating all of the RGB signals acquired from the three color sensors. Therefore, the proposed strategy was effective for improving sample discriminability. To further examine the dissimilar responses of each color sensor to odor molecules, typical odor components in the samples (allyl disulfide, allyl methyl disulfide, and perillaldehyde) were measured using each color sensor, and differences in RGB signals were analyzed. Full article
(This article belongs to the Section Chemical Sensors)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Wild, WHW, and 4E sensors containing the amino acid residues of glutamic acid (E)-glycine (G)-aspartic acid (D), tryptophan (W)-histidine (H)-tryptophan (W), and four repeating glutamic acids (E), respectively.</p> Full article ">Figure 2
<p>Overall experimental setup for measurement of samples using color sensors.</p> Full article ">Figure 3
<p>Average ΔRGB intensity of garlic samples measured in the three sections of Wild, WHW, and 4E sensors. The solid and dotted lines indicate the domestic and imported samples, respectively.</p> Full article ">Figure 4
<p>The same ΔRGB intensities shown in <a href="#sensors-21-00986-f003" class="html-fig">Figure 3</a> with standard deviation (shading) for the domestic (left plot) and imported samples (right plot).</p> Full article ">Figure 5
<p>Average ΔRGB intensity of allyl disulfide acquired using Wild, WHW, and 4E sensors. The molecular structure of allyl disulfide is also shown.</p> Full article ">Figure 6
<p>Variation of average ΔRGB intensity occurring in the three sections of Wild, WHW, and 4E sensors in the measurement of onion samples (solid line: domestic sample, dotted line: imported sample).</p> Full article ">Figure 7
<p>The same ΔRGB intensities shown in <a href="#sensors-21-00986-f006" class="html-fig">Figure 6</a> with standard deviation (shading) for the domestic (left plot) and imported samples (right plot).</p> Full article ">Figure 8
<p>Average ΔRGB intensity of allyl methyl disulfide individually acquired using Wild, WHW, and 4E sensors. The molecular structure of allyl methyl disulfide is also shown.</p> Full article ">Figure 9
<p>Average ΔRGB intensity of perilla samples measured in the three sections of Wild, WHW, and 4E sensors. The solid and dotted lines indicate the domestic and imported samples, respectively.</p> Full article ">Figure 10
<p>The same ΔRGB intensities shown in <a href="#sensors-21-00986-f009" class="html-fig">Figure 9</a> with standard deviation (shading) for the domestic (left plot) and imported samples (right plot).</p> Full article ">Figure 11
<p>Average ΔRGB intensity of perillaldehyde individually acquired using Wild, WHW, and 4E sensors. The molecular structure of perillaldehyde is also shown.</p> Full article ">Figure 12
<p>Average ΔRGB intensity of allyl disulfide, allyl methyl disulfide, and perillaldehyde measured using Wild (top), WHW (bottom), and 4E sensors (bottom).</p> Full article ">
10 pages, 1250 KiB  
Communication
Assessment of the Shank-to-Vertical Angle While Changing Heel Heights Using a Single Inertial Measurement Unit in Individuals with Incomplete Spinal Cord Injury Wearing an Ankle-Foot-Orthosis
by Lysanne A. F. de Jong, Yvette L. Kerkum, Tom de Groot, Marije Vos-van der Hulst, Ilse J. W. van Nes and Noel L. W. Keijsers
Sensors 2021, 21(3), 985; https://doi.org/10.3390/s21030985 - 2 Feb 2021
Cited by 2 | Viewed by 3785
Abstract
Previous research showed that an Inertial Measurement Unit (IMU) on the anterior side of the shank can accurately measure the Shank-to-Vertical Angle (SVA), which is a clinically-used parameter to guide tuning of ankle-foot orthoses (AFOs). However, in this context it is specifically important [...] Read more.
Previous research showed that an Inertial Measurement Unit (IMU) on the anterior side of the shank can accurately measure the Shank-to-Vertical Angle (SVA), which is a clinically-used parameter to guide tuning of ankle-foot orthoses (AFOs). However, in this context it is specifically important that differences in the SVA are detected during the tuning process, i.e., when adjusting heel height. This study investigated the validity of the SVA as measured by an IMU and its responsiveness to changes in AFO-footwear combination (AFO-FC) heel height in persons with incomplete spinal cord injury (iSCI). Additionally, the effect of heel height on knee flexion-extension angle and internal moment was evaluated. Twelve persons with an iSCI walked with their own AFO-FC in three different conditions: (1) without a heel wedge (refHH), (2) with 5 mm heel wedge (lowHH) and (3) with 10 mm heel wedge (highHH). Walking was recorded by a single IMU on the anterior side of the shank and a 3D gait analysis (3DGA) simultaneously. To estimate validity, a paired t-test and intraclass correlation coefficient (ICC) between the SVAIMU and SVA3DGA were calculated for the refHH. A repeated measures ANOVA was performed to evaluate the differences between the heel heights. A good validity with a mean difference smaller than 1 and an ICC above 0.9 was found for the SVA during midstance phase and at midstance. Significant differences between the heel heights were found for changes in SVAIMU (p = 0.036) and knee moment (p = 0.020) during the midstance phase and in SVAIMU (p = 0.042) and SVA3DGA (p = 0.006) at midstance. Post-hoc analysis revealed a significant difference between the ref and high heel height condition for the SVAIMU (p = 0.005) and knee moment (p = 0.006) during the midstance phase and for the SVAIMU (p = 0.010) and SVA3DGA (p = 0.006) at the instant of midstance. The SVA measured with an IMU is valid and responsive to changing heel heights and equivalent to the gold standard 3DGA. The knee joint angle and knee joint moment showed concomitant changes compared to SVA as a result of changing heel height. Full article
(This article belongs to the Special Issue Wearable Sensors for Movement Analysis)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>A</b>) Frontal view of the lower leg with additional markers on tibia tuberosity (TT) and the shank (SH1 and SH2). (<b>B</b>) Sagittal view of the lower leg with the corresponding coordinate system of the IMU (XYZimu). (<b>C</b>) Sagittal view of 5 mm heel wedge (lowHH). (<b>D</b>) Sagittal view of 10 mm heel wedge (highHH).</p> Full article ">Figure 2
<p>Mean SVA<sub>IMU</sub> [deg] (<b>A</b>), SVA<sub>3DGA</sub> [deg] (<b>B</b>), knee flexion-extension angle [deg] (<b>C</b>) and internal knee flexion-extension moment [Nm/kg] (<b>D</b>) during the gait cycle for refHH (green), lowHH (blue) and highHH (pink). The shaded area indicates the midstance phase (10–30%), the black line indicates the instant of midstance (34%) and the dashed line indicates toe off (68%).</p> Full article ">
12 pages, 734 KiB  
Article
The Simulated Characterization and Suitability of Semiconductor Detectors for Strontium 90 Assay in Groundwater
by Graeme Turkington, Kelum A. A. Gamage and James Graham
Sensors 2021, 21(3), 984; https://doi.org/10.3390/s21030984 - 2 Feb 2021
Cited by 1 | Viewed by 2708
Abstract
This paper examines the potential deployment of a 10 mm × 10 mm × 1 mm cadmium telluride detector for strontium-90 measurement in groundwater boreholes at nuclear decommissioning sites. Geant4 simulation was used to model the deployment of the detector in a borehole [...] Read more.
This paper examines the potential deployment of a 10 mm × 10 mm × 1 mm cadmium telluride detector for strontium-90 measurement in groundwater boreholes at nuclear decommissioning sites. Geant4 simulation was used to model the deployment of the detector in a borehole monitoring contaminated groundwater. It was found that the detector was sensitive to strontium-90, yttrium-90, caesium-137, and potassium-40 decay, some of the significant beta emitters found at Sellafield. However, the device showed no sensitivity to carbon-14 decay, due to the inability of the weak beta emission to penetrate both the groundwater and the detector shielding. The limit of detection for such a sensor when looking at solely strontium-90 decay was calculated as 323 BqL−1 after a 1-h measurement and 66 BqL−1 after a 24-h measurement. A gallium-arsenide (GaAs) sensor with twice the surface area, but 0.3% of the thickness was modelled for comparison. Using this sensor, sensitivity was increased, such that the limit of detection for strontium-90 was 91 BqL−1 after 1 h and 18 BqL−1 after 24 h. However, this sensor sacrifices the potential to identify the present radionuclides by their end-point energy. Additionally, the feasibility of using flexible detectors based on solar cell designs to maximise the surface area of detectors has been modelled. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>The simulation scenario illustrated. The detector is deployed to the surface of contaminated groundwater. Decaying radionuclides are randomly distributed throughout the water.</p> Full article ">Figure 2
<p><math display="inline"><semantics> <msup> <mrow/> <mn>90</mn> </msup> </semantics></math>Sr counted in a CdTe detector in a groundwater borehole simulation at different activities.</p> Full article ">Figure 3
<p><math display="inline"><semantics> <msup> <mrow/> <mn>90</mn> </msup> </semantics></math>Sr and <math display="inline"><semantics> <msup> <mrow/> <mn>90</mn> </msup> </semantics></math>Y counts recorded in a CdTe detector in a groundwater borehole simulation. The combined spectrum is plotted in yellow.</p> Full article ">Figure 4
<p><math display="inline"><semantics> <msup> <mrow/> <mn>90</mn> </msup> </semantics></math>Sr and <math display="inline"><semantics> <msup> <mrow/> <mn>90</mn> </msup> </semantics></math>Y counts recorded in a GaAs detector in a groundwater borehole simulation. The combined spectrum is plotted in yellow.</p> Full article ">Figure 5
<p>A flexible CdTe detector, white rectangle, applied to the surface of a cylindrical detector, as depicted in the Geant4 simulation with decaying <math display="inline"><semantics> <msup> <mrow/> <mn>90</mn> </msup> </semantics></math>Sr, yellow dots, and beta particles, red lines. Some surfaces have been hidden for clarity.</p> Full article ">
17 pages, 425 KiB  
Article
Efficient Wind Speed Forecasting for Resource-Constrained Sensor Devices
by Sergio Herrería-Alonso, Andrés Suárez-González, Miguel Rodríguez-Pérez, Raúl F. Rodríguez-Rubio and Cándido López-García
Sensors 2021, 21(3), 983; https://doi.org/10.3390/s21030983 - 2 Feb 2021
Cited by 4 | Viewed by 2410
Abstract
Wind energy harvesting technology is one of the most popular power sources for wireless sensor networks. However, given its irregular nature, wind energy availability experiences significant variations and, therefore, wind-powered devices need reliable forecasting models to effectively adjust their energy consumption to the [...] Read more.
Wind energy harvesting technology is one of the most popular power sources for wireless sensor networks. However, given its irregular nature, wind energy availability experiences significant variations and, therefore, wind-powered devices need reliable forecasting models to effectively adjust their energy consumption to the dynamics of energy harvesting. On the other hand, resource-constrained devices with limited hardware capacities (such as sensor nodes) must resort to forecasting schemes of low complexity for their predictions in order to avoid squandering their scarce power and computing capabilities. In this paper, we present a new efficient ARIMA-based forecasting model for predicting wind speed at short-term horizons. The performance results obtained using real data sets show that the proposed ARIMA model can be an excellent choice for wind-powered sensor nodes due to its potential for achieving accurate enough predictions with very low computational burden and memory overhead. In addition, it is very simple to setup, since it can dynamically adapt to varying wind conditions and locations without requiring any particular reconfiguration or previous data training phase for each different scenario. Full article
(This article belongs to the Special Issue Green Sensors Networking)
►▼ Show Figures

Figure 1

Figure 1
<p>Block diagram of a wind-powered energy harvesting (EH) node. Solid (dotted) lines represent energy (data) transfer.</p> Full article ">Figure 2
<p>Autocorrelation functions (ACFs) for <math display="inline"><semantics> <mrow> <mi>d</mi> <mo>=</mo> <mn>0</mn> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <mi>d</mi> <mo>=</mo> <mn>1</mn> </mrow> </semantics></math>.</p> Full article ">Figure 3
<p>Partial autocorrelation functions (PACFs) for <math display="inline"><semantics> <mrow> <mi>d</mi> <mo>=</mo> <mn>1</mn> </mrow> </semantics></math>.</p> Full article ">Figure 4
<p>Relative mean absolute error (MAE) difference (persistence model used as baseline).</p> Full article ">Figure 5
<p>Percentage of optimistic predictions.</p> Full article ">
30 pages, 17538 KiB  
Article
Fluid–Structure Coupling Effects in a Dual U-Tube Coriolis Mass Flow Meter
by Yuh-Chung Hu, Zen-Yu Chen and Pei-Zen Chang
Sensors 2021, 21(3), 982; https://doi.org/10.3390/s21030982 - 2 Feb 2021
Cited by 13 | Viewed by 6265
Abstract
Coriolis mass flowmeters are highly customized products involving high-degree fluid-structure coupling dynamics and high-precision manufacture. The typical delay from from order to shipment is at least 4 months. This paper presents some important design considerations through simulation and experiments, so as to provide [...] Read more.
Coriolis mass flowmeters are highly customized products involving high-degree fluid-structure coupling dynamics and high-precision manufacture. The typical delay from from order to shipment is at least 4 months. This paper presents some important design considerations through simulation and experiments, so as to provide manufacturers with a more time-efficient product design and manufacture process. This paper aims at simulating the fluid-structure coupling dynamics of a dual U-tube Coriolis mass flowmeter through the COMSOL simulation package. The simulation results are experimentally validated using a dual U-tube CMF manufactured by Yokogawa Co., Ltd. in a TAF certified flow testing factory provided by FineTek Co., Ltd. Some important design considerations are drawn from simulation and experiment. The zero drift will occur when the dual U-tube structure is unbalanced and therefore the dynamic balance is very important in the manufacturing of dual U-tube CMF. The fluid viscosity can be determined from the driving current of the voice coil actuator or the pressure loss between the inlet and outlet of CMF. Finally, the authors develop a simulation application based on COMSOL’s development platform. Users can quickly evaluate their design through by using this application. The present application can significantly shorten product design and manufacturing time. Full article
(This article belongs to the Section Physical Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Sample CMF manufactured by Yokogawa Co., Ltd.</p> Full article ">Figure 2
<p>Geometric structure of the CMF.</p> Full article ">Figure 3
<p>Density experiment setup: (<b>a</b>) the demonstration mass flow meter in a constant temperature chamber; (<b>b</b>) the densitometer manufactured by Kyoto Electronics Manufacturing Co., Ltd.</p> Full article ">Figure 4
<p>Mass flow rate experiment: (<b>a</b>) the experiment setup; (<b>b</b>) the schematic of experiment setup.</p> Full article ">Figure 5
<p>Schematic and photo of the installation of a dual U-tube CMF with a deflection angle.2. Simulation and Experiment.</p> Full article ">Figure 6
<p>Setup of the structural imbalance experiment: (<b>a</b>) mass added 10 mm below only Sensor 1, (<b>b</b>) mass added 10 mm below only Sensor 2, and (<b>c</b>) an oscilloscope used to capture sensor output signals.</p> Full article ">Figure 7
<p>Simulation steps for a CMF.</p> Full article ">Figure 8
<p>The physical model of the CMF.</p> Full article ">Figure 9
<p>The schematic of turbulent flow field.</p> Full article ">Figure 10
<p>The mesh of flow field: (<b>a</b>) the region of boundary layer mesh; (<b>b</b>) the boundary layer mesh.</p> Full article ">Figure 11
<p>The convergence analysis of the boundary mesh.</p> Full article ">Figure 12
<p>Density–resonance frequency relationship in the simulation and experiment.</p> Full article ">Figure 13
<p>Mass flow rate experiment results: (<b>a</b>) the time difference of output signals vs. the mass flow rate; (<b>b</b>) the relative error analysis of linear regression; (<b>c</b>) the average relative error analysis of linear regression; (<b>d</b>) the CMF measurement repeatability analysis.</p> Full article ">Figure 14
<p>Mass flow rate–phase difference relationship in the simulation and experiment.</p> Full article ">Figure 15
<p>The influence of installation with a deflection angle on: (<b>a</b>) CMF resonance frequency; (<b>b</b>) CMF sensitivity.</p> Full article ">Figure 16
<p>The raw output voltage signals of the two motion sensors in the first 15 ms: (<b>a</b>) no additional mass, (<b>b</b>) with an additional mass below Sensor 1, and (<b>c</b>) with an additional mass below Sensor 2.</p> Full article ">Figure 17
<p>COMSOL-based simulation of displacement signals when a 10-g mass is added below (<b>a</b>) Sensor 1 and (<b>b</b>) Sensor 2.</p> Full article ">Figure 18
<p>The mass flow–phase difference relationship in an imbalanced CMF structure: (<b>a</b>) the simulation results; (<b>b</b>) the experimental results.</p> Full article ">Figure 19
<p>COMSOL-based simulation of the pressure drop of a dual U-tube CMF.</p> Full article ">Figure 20
<p>Schematic of the bend.</p> Full article ">Figure 21
<p>Relationship between fluid viscosity and the pipe flow pressure drop.</p> Full article ">Figure 22
<p>The voice coil actuator/sensor: (<b>a</b>) the photo; (<b>b</b>) the schematic.</p> Full article ">Figure 23
<p>Simulation of fluid viscosity vs. driving force when the flow rate was 1 kg/s: (<b>a</b>) the simulation result. (<b>b</b>) the cutoff point of linear and nonlinear region was at <span class="html-italic">Re</span> = 40,000. (<b>c</b>) the linear function by curve fitting as <span class="html-italic">Re</span> &lt; 40,000. (<b>d</b>) the cubic polynomial function by curve fitting as <span class="html-italic">Re</span> &gt; 40,000.</p> Full article ">Figure 24
<p>Simulation of fluid viscosity vs. driving force when the mass flow rate was 0.8 kg/s: (<b>a</b>) the cutoff point of linear and nonlinear region was at <span class="html-italic">Re</span> = 40,000; (<b>b</b>) the linear function by curve fitting as <span class="html-italic">Re</span> &lt; 40,000; (<b>c</b>) the cubic polynomial function by curve fitting as <span class="html-italic">Re</span> &gt; 40,000.</p> Full article ">Figure 25
<p>Simulation of fluid viscosity vs. driving force when the mass flow rate was 1.2 kg/s: (<b>a</b>) the cutoff point of linear and nonlinear region was at <span class="html-italic">Re</span> = 40,000; (<b>b</b>) the linear function by curve fitting as <span class="html-italic">Re</span> &lt; 40,000; (<b>c</b>) the cubic polynomial function by curve fitting as <span class="html-italic">Re</span> &gt; 40,000.</p> Full article ">Figure 26
<p>The flowchart of deducing the fluid viscosity from the driving force.</p> Full article ">Figure 27
<p>The influence of motion sensor position on: (<b>a</b>) the vibration amplitude at the position of motion sensor; (<b>b</b>) the phase difference to the output signals of the motion sensors.</p> Full article ">Figure 28
<p>Cross-sectional schematic of the conical splitter structure.</p> Full article ">Figure 29
<p>Simulation results of the conical splitter structure.</p> Full article ">Figure 30
<p>Taper curved conical splitter structure: (<b>a</b>) isometric and (<b>b</b>) front views.</p> Full article ">Figure A1
<p>The simple introduction page of the App.</p> Full article ">Figure A2
<p>The geometry establishment and resonance frequency simulation page of the App.</p> Full article ">Figure A3
<p>Message window: (<b>a</b>) for geometry creation and successful solution; (<b>b</b>) for reminding the parameters have been changed.</p> Full article ">Figure A4
<p>The fluid dynamics and simulation of the fluid-structure interaction page of the App.</p> Full article ">Figure A5
<p>Simulation results imported from the COMSOL app into an MS Excel spreadsheet.</p> Full article ">
23 pages, 1543 KiB  
Article
A-WEAR Bracelet for Detection of Hand Tremor and Bradykinesia in Parkinson’s Patients
by Asma Channa, Rares-Cristian Ifrim, Decebal Popescu and Nirvana Popescu
Sensors 2021, 21(3), 981; https://doi.org/10.3390/s21030981 - 2 Feb 2021
Cited by 38 | Viewed by 6185
Abstract
Parkinson’s disease patients face numerous motor symptoms that eventually make their life different from those of normal healthy controls. Out of these motor symptoms, tremor and bradykinesia, are relatively prevalent in all stages of this disease. The assessment of these symptoms is usually [...] Read more.
Parkinson’s disease patients face numerous motor symptoms that eventually make their life different from those of normal healthy controls. Out of these motor symptoms, tremor and bradykinesia, are relatively prevalent in all stages of this disease. The assessment of these symptoms is usually performed by traditional methods where the accuracy of results is still an open question. This research proposed a solution for an objective assessment of tremor and bradykinesia in subjects with PD (10 older adults aged greater than 60 years with tremor and 10 older adults aged greater than 60 years with bradykinesia) and 20 healthy older adults aged greater than 60 years. Physical movements were recorded by means of an AWEAR bracelet developed using inertial sensors, i.e., 3D accelerometer and gyroscope. Participants performed upper extremities motor activities as adopted by neurologists during the clinical assessment based on Unified Parkinson’s Disease Rating Scale (UPDRS). For discriminating the patients from healthy controls, temporal and spectral features were extracted, out of which non-linear temporal and spectral features show greater difference. Both supervised and unsupervised machine learning classifiers provide good results. Out of 40 individuals, neural net clustering discriminated 34 individuals in correct classes, while the KNN approach discriminated 91.7% accurately. In a clinical environment, the doctor can use the device to comprehend the tremor and bradykinesia of patients quickly and with higher accuracy. Full article
(This article belongs to the Special Issue Applications and Innovations on Sensor-Enabled Wearable Devices)
►▼ Show Figures

Figure 1

Figure 1
<p>Block design for the bracelet.</p> Full article ">Figure 2
<p>Components used for development of the AWEAR bracelet.</p> Full article ">Figure 3
<p>AWEAR bracelet developmental stages.</p> Full article ">Figure 4
<p>Illustration of the whole process for detection of tremor and bradykinesia.</p> Full article ">Figure 5
<p>Neural network architecture.</p> Full article ">Figure 6
<p>Visualization of healthy control signals before and after filtering in time and frequency domain.</p> Full article ">Figure 7
<p>Visualization of PD patient signals before and after filtering in time and frequency domain.</p> Full article ">Figure 8
<p>The histograms of the extracted features.</p> Full article ">Figure 8 Cont.
<p>The histograms of the extracted features.</p> Full article ">Figure 9
<p>Features sorted by importance for the classifiers. The right side displays the ANOVA results, whereas the bars from the left side depict the normalized scores of different features.</p> Full article ">Figure 10
<p>Results of the trained model.</p> Full article ">Figure 11
<p>KNN classifier results.</p> Full article ">Figure 12
<p>ROC curves.</p> Full article ">
17 pages, 5428 KiB  
Article
Intelligent Control Method of Hoisting Prefabricated Components Based on Internet-of-Things
by Yuhong Zhao, Cunfa Cao, Zhansheng Liu and Enyi Mu
Sensors 2021, 21(3), 980; https://doi.org/10.3390/s21030980 - 2 Feb 2021
Cited by 16 | Viewed by 3251
Abstract
Prefabricated buildings are widely used because of their green environmental protection and high degree of industrialization. However, in construction process, there are some defects such as small wireless network coverage, high-energy consumption, inaccurate control, and backward blind hoisting methods in the hoisting process [...] Read more.
Prefabricated buildings are widely used because of their green environmental protection and high degree of industrialization. However, in construction process, there are some defects such as small wireless network coverage, high-energy consumption, inaccurate control, and backward blind hoisting methods in the hoisting process of prefabricated components (PC). Internet-of-Things (IoT) technology can be used to collect and transmit data to strengthen the management of construction sites. The purpose of this study was to establish an intelligent control method in the construction and hoisting process of PC by using IoT technology. Long Range Radio (LoRa) technology was used to conduct data terminal acquisition and wireless transmission in the construction site. The Inertial Measurement Unit (IMU), Global Positioning System (GPS), and other multi-sensor fusion was used to collect information during the hoisting process of PC, and multi-sensor information was fused by fusion location algorithm for location control. Finally, the feasibility of this method was verified by a project as a case. The results showed that the IoT technology can strengthen the management ability of PC in the hoisting process, and improve the visualization level of the hoisting process of PC. Analysis of the existing outdated PC hoisting management methods, LoRa, IMU, GPS and other sensors were used for data acquisition and transmission, the PC hoisting multi-level management and intelligent control. Full article
(This article belongs to the Special Issue LoRa Sensor Network)
►▼ Show Figures

Figure 1

Figure 1
<p>LoRaWAN network architecture.</p> Full article ">Figure 2
<p>Application functional architecture.</p> Full article ">Figure 3
<p>Logic diagram of the hoisting control method.</p> Full article ">Figure 4
<p>Logical diagram of the positioning algorithm.</p> Full article ">Figure 5
<p>Building Information Modeling (BIM) model for field layout of construction site.</p> Full article ">Figure 6
<p>LoRa transmission terminal.</p> Full article ">Figure 7
<p>LoRa gateway installed on the construction site.</p> Full article ">Figure 8
<p>Analysis of power consumption of positioning terminal based on the LoRa network.</p> Full article ">Figure 9
<p>Analysis of data transmission rate and data transmission time in the LoRa network.</p> Full article ">Figure 10
<p>Control flow of PC hoisting.</p> Full article ">Figure 11
<p>Hoisting track diagram of PC.</p> Full article ">
31 pages, 11577 KiB  
Review
Recent Advances in Aptamer Sensors
by Samy M. Shaban and Dong-Hwan Kim
Sensors 2021, 21(3), 979; https://doi.org/10.3390/s21030979 - 2 Feb 2021
Cited by 64 | Viewed by 11869
Abstract
Recently, aptamers have attracted attention in the biosensing field as signal recognition elements because of their high binding affinity toward specific targets such as proteins, cells, small molecules, and even metal ions, antibodies for which are difficult to obtain. Aptamers are single oligonucleotides [...] Read more.
Recently, aptamers have attracted attention in the biosensing field as signal recognition elements because of their high binding affinity toward specific targets such as proteins, cells, small molecules, and even metal ions, antibodies for which are difficult to obtain. Aptamers are single oligonucleotides generated by in vitro selection mechanisms via the systematic evolution of ligand exponential enrichment (SELEX) process. In addition to their high binding affinity, aptamers can be easily functionalized and engineered, providing several signaling modes such as colorimetric, fluorometric, and electrochemical, in what are known as aptasensors. In this review, recent advances in aptasensors as powerful biosensor probes that could be used in different fields, including environmental monitoring, clinical diagnosis, and drug monitoring, are described. Advances in aptamer-based colorimetric, fluorometric, and electrochemical aptasensing with their advantages and disadvantages are summarized and critically discussed. Additionally, future prospects are pointed out to facilitate the development of aptasensor technology for different targets. Full article
(This article belongs to the Special Issue Recent Advances in Apta-Biosensors)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>A</b>) Schematic illustration of the colorimetric aptamer sensor for <span class="html-italic">E. coli</span> assay via target inducing aggregation of gold nanoparticles (AuNPs), reproduced from [<a href="#B84-sensors-21-00979" class="html-bibr">84</a>]. (<b>B</b>) Schematic illustration of the colorimetric aptasensor for Cd<sup>2+</sup> assay based on AuNP aggregation, reproduced from [<a href="#B86-sensors-21-00979" class="html-bibr">86</a>]. (<b>C</b>) Schematic procedure for an <span class="html-italic">S. typhimurium</span> bacterium assay via a proposed colorimetric SMBs-Apt1 sandwich strategy, reproduced from [<a href="#B87-sensors-21-00979" class="html-bibr">87</a>]. (<b>D</b>) Schematic illustration of a colorimetric aptasensor assay for AFM1 based on salt aggregation of AuNPs induced by AuNP–aptamer competitive binding, reproduced from [<a href="#B89-sensors-21-00979" class="html-bibr">89</a>].</p> Full article ">Figure 2
<p>(<b>A</b>) Schematic illustration of AFB1 assay with the label-free colorimetric aptasensor using the competition between the specific aptamer, CPP, and AuNPs inducing AuNP aggregation, reproduced from [<a href="#B90-sensors-21-00979" class="html-bibr">90</a>]. (<b>B</b>) Schematic protocol of the dopamine assay via aptamer-assisted AuNP-induced NaCl salt aggregation with two possible mechanisms of sensing, reproduced from [<a href="#B91-sensors-21-00979" class="html-bibr">91</a>]. (<b>C</b>) Schemes of the As (III) assay based on the aptamer–As (III) competitive interaction inducing salt–AuNP aggregation, reproduced from [<a href="#B92-sensors-21-00979" class="html-bibr">92</a>]. (<b>D</b>) Schematic principle of <span class="html-italic">S. typhimurium</span> assay based on the peroxidase-mimicking activity of dual aptamers@BSA-AuNCs probes inducing a blue color for 3,3′,5,5′-tetramethylbenzidine (TMB), reproduced from [<a href="#B97-sensors-21-00979" class="html-bibr">97</a>].</p> Full article ">Figure 3
<p>(<b>A</b>) Schematic illustration of the cortisol assay based on localized surface plasmon resonance (LSPR) aptasensor, reproduced from [<a href="#B99-sensors-21-00979" class="html-bibr">99</a>]. (<b>B</b>) Scheme representing the thrombin assay via sandwich colorimetric solid-phase aptasensor based on the enhanced LSPR, reproduced from [<a href="#B100-sensors-21-00979" class="html-bibr">100</a>].</p> Full article ">Figure 4
<p>(<b>A</b>) Principal strategy for Cd<sup>2+</sup> assay via the enhanced peroxidase-like activity of Au–MoS<sub>2</sub>-based aptamer, reproduced from [<a href="#B101-sensors-21-00979" class="html-bibr">101</a>]. (<b>B</b>) Principal protocol for the PDGF-BB assay via a glucose oxidase enzyme, inducing changes in pH-equipped immunomagnetic beads and mb-RCA as a signal amplifier, reproduced from [<a href="#B103-sensors-21-00979" class="html-bibr">103</a>]. (<b>C</b>) Schematic diagram showing the procedure and mechanism of the ATP assay via colorimetric aptamer technology based on the peroxidase-like catalysis properties of Fe<sub>3</sub>O<sub>4</sub>, reproduced from [<a href="#B106-sensors-21-00979" class="html-bibr">106</a>]. (<b>D</b>) Colorimetric procedure for Pb<sup>2+</sup> assay based on the peroxidase-mimicking activity of graphene/Fe<sub>3</sub>O<sub>4</sub>-AuNPs aptamer technology, reproduced from [<a href="#B107-sensors-21-00979" class="html-bibr">107</a>].</p> Full article ">Figure 5
<p>(<b>A</b>) Schematic illustration of the ELISA for assaying of 17β-E2 with an assisting aptamer–antibody sandwich functionalized with AuNPs, reproduced from [<a href="#B109-sensors-21-00979" class="html-bibr">109</a>]. (<b>B</b>) Principal assay procedures for mycotoxins using a green ELISA based on a single-stranded binding protein (SSB)-assisted aptamer; the SSB was dispensed into polystyrene 96 plates; reproduced from [<a href="#B110-sensors-21-00979" class="html-bibr">110</a>]. (<b>C</b>) Schematic illustration outlining the multicolor and photothermal assay for prostate-specific antigen (PSA) via magnetic beads assisting aptamer separation, reproduced from [<a href="#B111-sensors-21-00979" class="html-bibr">111</a>]. (<b>D</b>) Detection of <span class="html-italic">E.coli O157:H7</span> in milk via the developed naked-eye EA-Sensor, reproduced from [<a href="#B112-sensors-21-00979" class="html-bibr">112</a>].</p> Full article ">Figure 6
<p>(<b>A</b>) Schematics showing the preparation of PAA@Arg@ATT-AuNC as a fluorescent probe for trichothecenes A (T-2) toxin assay, reproduced from [<a href="#B119-sensors-21-00979" class="html-bibr">119</a>]. (<b>B</b>) Graphical illustration for the assaying of digoxin (DGX) using the aptamer/AuNPs/g-C3N4NS sensor probe, reproduced from [<a href="#B120-sensors-21-00979" class="html-bibr">120</a>]. (<b>C</b>) Schematic illustration of the zearalenone assay using a ratiometric fluorescent nanoprobe with dual emission at 518 and 608 nm, reproduced from [<a href="#B121-sensors-21-00979" class="html-bibr">121</a>]. (<b>D</b>) Schematic description of the protocol employed for a patulin (PAT) assay via CA-MWCNTs) with quenching of aptamer-tagged carboxyfluorescein, reproduced from [<a href="#B130-sensors-21-00979" class="html-bibr">130</a>]. (<b>E</b>) Schematic description of AFB1 detection based on UiO-66-NH<sub>2</sub> and aptamer-functionalized TAMRA dye, reproduced from [<a href="#B123-sensors-21-00979" class="html-bibr">123</a>]. (<b>F</b>) Schematic of the fluorescent aptasensor for an AMP assay based on the competitive quenching between the AMP aptamer complementary strand and AuNPs, reproduced from [<a href="#B131-sensors-21-00979" class="html-bibr">131</a>]. (<b>G</b>) Schematic showing an ultrasensitive fluorometric aptasensor for IFN-γ detection by dual atom transfer radical polymerization (ATRP) amplification, reproduced from [<a href="#B132-sensors-21-00979" class="html-bibr">132</a>]. (<b>H</b>) Schematic showing the fabrication of PDA–Apt liposomes and fluorescence imaging of plasma membrane glycoprotein MUC1, reproduced from [<a href="#B133-sensors-21-00979" class="html-bibr">133</a>]. (<b>I</b>) Schematic description of isocarbophos assay via fluorometric aptsensor based on AT-rich three-way junction DNA template copper nanoparticles and Fe<sub>3</sub>O<sub>4</sub>@GO, reproduced from [<a href="#B134-sensors-21-00979" class="html-bibr">134</a>].</p> Full article ">Figure 7
<p>Schematic description of the (<b>A</b>) sensing platform for <span class="html-italic">E. coli</span> detection using fluorescent UCNP–WS2 nanosheet, reproduced from [<a href="#B140-sensors-21-00979" class="html-bibr">140</a>]. (<b>B</b>) Proposed fluorescence aptasensor for <span class="html-italic">E. coli.</span> using MNP–aptamer–cDNA–UCNP probe, reproduced from [<a href="#B141-sensors-21-00979" class="html-bibr">141</a>]. (<b>C</b>) Fluorometric aptasensor for malathion assay based on the FRET between UCNPs and GNPs, reproduced from [<a href="#B142-sensors-21-00979" class="html-bibr">142</a>]. (<b>D</b>) Proposed scheme of ATP assay based on ratiometric fluorescence from the binding between the ATP and aptamer complexes, reproduced from [<a href="#B145-sensors-21-00979" class="html-bibr">145</a>].</p> Full article ">Figure 8
<p>(<b>A</b>) Schematic illustration of the ATP assay via quenching of DNA/AgNCs by AuNRs functionalized with aptamer and Exo III for signal amplification, reproduced from [<a href="#B152-sensors-21-00979" class="html-bibr">152</a>]. (<b>B</b>) Fluorometric assay for acetamiprid using exonuclease integrated with the aptamer protocol, reproduced from [<a href="#B154-sensors-21-00979" class="html-bibr">154</a>]. (<b>C</b>) Exosomal assay based on the AIE mechanism using graphene oxide (GO) as a quencher for tertiary amine-containing tetraphenylethene (TPE-TA) dye, reproduced from [<a href="#B121-sensors-21-00979" class="html-bibr">121</a>]. (<b>D</b>) Dual-mode fluorescent aptasensor using both aptamer and a DNAzyme to assay ATP with two different mechanisms, fluorometric and colorimetric signals, reproduced from [<a href="#B159-sensors-21-00979" class="html-bibr">159</a>]. (<b>E</b>) TTX assay based on the difference in fluorescence response of the berberine reporter, reproduced from [<a href="#B160-sensors-21-00979" class="html-bibr">160</a>]. (<b>F</b>) Turn-on fluorescence aptasensor assay for chloramphenicol based on oligomer quencher release, reproduced from [<a href="#B163-sensors-21-00979" class="html-bibr">163</a>].</p> Full article ">Figure 9
<p>(<b>A</b>–<b>D</b>) Schematic illustrations for the fabrication of electrochemical aptasensors. (<b>A</b>) Impedimetric assay using glassy carbon electrode (GCE) immobilized with OTA aptamer, reproduced from [<a href="#B190-sensors-21-00979" class="html-bibr">190</a>]. (<b>B</b>) Impedimetric assay for amoxicillin based on the synergetic effect between the amoxicillin and its aptamer on the surface of the GCE coated with TiO<sub>2</sub>-g-C<sub>3</sub>N<sub>4</sub>-AuNPs, reproduced from [<a href="#B191-sensors-21-00979" class="html-bibr">191</a>]. (<b>C</b>) A sandwich-type aptasensor for thrombin assay using CSPH hydrogel-modified electrode, reproduced from [<a href="#B192-sensors-21-00979" class="html-bibr">192</a>]. (<b>D</b>) PSA assay based on polyaniline (PANI)/AuNPs as a conducting layer on GCE immobilized with a PSA aptamer, reproduced from [<a href="#B193-sensors-21-00979" class="html-bibr">193</a>].</p> Full article ">Figure 10
<p>(<b>A</b>–<b>C</b>) Schematic protocols for electrochemical aptasensor based on the target-responsive label electroactive dye amplification strategy. (<b>A</b>) OTC assay using AuNP/cMWCNT/cDNA@thionine probe as a signaler releasing thionine, reproduced from [<a href="#B196-sensors-21-00979" class="html-bibr">196</a>]. (<b>B</b>) PAT assay using MOF@Methylene blue signal tags, with signals as a function of PAT concentrations, reproduced from [<a href="#B43-sensors-21-00979" class="html-bibr">43</a>]. (<b>C</b>) Thrombin assay based on target-responsive methylene blue (MB) release from the ZIF-8 surface as a function of the thrombin concentration, reproduced from [<a href="#B13-sensors-21-00979" class="html-bibr">13</a>]. (<b>D</b>) Schematic protocols for an electrochemical aptasensor based on an enzymatic catalytic reaction, reproduced from [<a href="#B201-sensors-21-00979" class="html-bibr">201</a>].</p> Full article ">Figure 11
<p>(<b>A</b>–<b>C</b>) Schematic protocols for electrochemical aptasensor integrated with signal amplification based on nuclease strategy. (<b>A</b>) Assaying of sulfadimethoxine using gold electrode fabricated with dsDNA consisting of DNA probe and specific aptamer for sulfadimethoxine inhibiting nuclease digestion, reproduced from [<a href="#B203-sensors-21-00979" class="html-bibr">203</a>]. (<b>B</b>) Malathion assay with the assistance of Exo I as a signal enhancer and layer of PDA-AuNPs deposited on GCE as an electrical conductivity enhancer, reproduced from [<a href="#B204-sensors-21-00979" class="html-bibr">204</a>]. (<b>C</b>) OTA assay using TiO<sub>2</sub> nanotube electrode functionalized with DNA probe tagged with MB, reproduced from [<a href="#B205-sensors-21-00979" class="html-bibr">205</a>]. (<b>D</b>) Schematic protocols for photoelectrochemical aptasensor ATZ assay with nuclease signal amplification using TiO<sub>2</sub> nanotube covered by a layer of AuNPs, further functionalized with ATZ aptamer/graphene mixture as a sensor probe, reproduced from [<a href="#B206-sensors-21-00979" class="html-bibr">206</a>].</p> Full article ">
18 pages, 13913 KiB  
Article
A 3D Informational Database for Automatic Archiving of Archaeological Pottery Finds
by Luca Di Angelo, Paolo Di Stefano, Emanuele Guardiani and Anna Eva Morabito
Sensors 2021, 21(3), 978; https://doi.org/10.3390/s21030978 - 2 Feb 2021
Cited by 18 | Viewed by 3647
Abstract
From archaeological excavations, huge quantities of material are recovered, usually in the form of fragments. Their correct interpretation and classification are laborious and time-consuming and requires measurement, analysis and comparison of several items. Basing these activities on quantitative methods that process 3D digital [...] Read more.
From archaeological excavations, huge quantities of material are recovered, usually in the form of fragments. Their correct interpretation and classification are laborious and time-consuming and requires measurement, analysis and comparison of several items. Basing these activities on quantitative methods that process 3D digital data from experimental measurements allows optimizing the entire restoration process, making it faster, more accurate and cheaper. The 3D point clouds, captured by the scanning process, are raw data that must be properly processed to be used in automatic systems for the analysis of archeological finds. This paper focuses on the integration of a shape feature recognizer, able to support the semantic decomposition of the ancient artifact into archaeological features, with a structured database, able to query the large amount of information extracted. Through the automatic measurement of the dimensional attributes of the various features, it is possible to facilitate the comparative analyses between archaeological artifacts and the inferences of the archaeologist and to reduce the routine work. Here, a dedicated database has been proposed, able to store the information extracted from huge quantities of archaeological material using a specific shape feature recognizer. This information is useful for making comparisons but also to improve the archaeological knowledge. The database has been implemented and used for the identification of pottery fragments and the reconstruction of archaeological vessels. Reconstruction, in particular, often requires the solution of complex problems, especially when it involves types of potsherds that cannot be treated with traditional methods. Full article
(This article belongs to the Special Issue Electrical and Electromagnetic Sensors and Methods in Archaeology and Monumental Heritage)
►▼ Show Figures

Figure 1

Figure 1
<p>UML diagram of the database.</p> Full article ">Figure 2
<p>Feature recognizer service.</p> Full article ">Figure 3
<p>Application component diagram.</p> Full article ">Figure 4
<p>JQuery AJAX library description.</p> Full article ">Figure 5
<p>Web interface of search page.</p> Full article ">Figure 6
<p>Web interface of object viewer. (<b>a</b>) The internal wall has been selected as displayed feature. (<b>b</b>) The rim is displayed.</p> Full article ">Figure 7
<p>Four screenshots of the application and some recognized features in a type-A sherd.</p> Full article ">Figure 8
<p>CRS segmentation of an archaeological pot. The different colors are used to highlight the non-adjacent CRS features [<a href="#B4-sensors-21-00978" class="html-bibr">4</a>].</p> Full article ">Figure 9
<p>Main geometrical and dimensional features recognized in an archeological sherd.</p> Full article ">Figure 10
<p>The collection of potsherds from Alba Fucens excavation.</p> Full article ">Figure 11
<p>The flowchart of the algorithm.</p> Full article ">Figure 12
<p>Software interface for pottery reconstruction: matching of the fragments’ profiles.</p> Full article ">Figure 13
<p>(<b>a</b>) The simulated assembly of the four archaeological fragments identified by the algorithm implemented. (<b>b</b>) The longitudinal profile and (<b>c</b>) the rendering of 3D virtual reconstruction of the archaeological vessel.</p> Full article ">
11 pages, 3523 KiB  
Communication
Gradiometer Using Separated Diamond Quantum Magnetometers
by Yuta Masuyama, Katsumi Suzuki, Akira Hekizono, Mitsuyasu Iwanami, Mutsuko Hatano, Takayuki Iwasaki and Takeshi Ohshima
Sensors 2021, 21(3), 977; https://doi.org/10.3390/s21030977 - 2 Feb 2021
Cited by 11 | Viewed by 4988
Abstract
The negatively charged nitrogen-vacancy (NV) center in diamonds is known as the spin defect and using its electron spin, magnetometry can be realized even at room temperature with extremely high sensitivity as well as a high dynamic range. However, a magnetically shielded enclosure [...] Read more.
The negatively charged nitrogen-vacancy (NV) center in diamonds is known as the spin defect and using its electron spin, magnetometry can be realized even at room temperature with extremely high sensitivity as well as a high dynamic range. However, a magnetically shielded enclosure is usually required to sense weak magnetic fields because environmental magnetic field noises can disturb high sensitivity measurements. Here, we fabricated a gradiometer with variable sensor length that works at room temperature using a pair of diamond samples containing negatively charged NV centers. Each diamond is attached to an optical fiber to enable free sensor placement. Without any magnetically shielding, our gradiometer realizes a magnetic noise spectrum comparable to that of a three-layer magnetically shielded enclosure, reducing the noises at the low-frequency range below 1 Hz as well as at the frequency of 50 Hz (power line frequency) and its harmonics. These results indicate the potential of highly sensitive magnetic sensing by the gradiometer using the NV center for applications in noisy environments such as outdoor and in vehicles. Full article
(This article belongs to the Special Issue Magnetic Sensing/Functionalized Devices and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Schematic representation of the gradiometer setup. A longer base length than the depth enabled us to selectively detect the target signal. (<b>b</b>) A schematic setup of the gradiometer using the nitrogen-vacancy (NV) centers in a pair of diamonds. A dichroic mirror selectively reflected the laser beam and transmitted the fluorescence from NV centers. The fluorescence from each diamond (sensor 1 and sensor 2) was detected by the respective photodiode connected to an oscilloscope.</p> Full article ">Figure 2
<p>Typical optically detected magnetic resonance (ODMR) spectra of the spatially isolated diamond samples of sensor 1 (Ch. 1) and sensor 2 (Ch. 2). The detected fluorescence at each detector is about 0.2 mW. The ODMR signals (dips) of the left and right sides correspond to magnetic sublevels m<sub>s</sub> = −1 and +1, respectively, for the NV center with [111] direction. The resonance of the other three of the four orientation axes is degenerate. Our microwave circuit drives the NV center with [111] direction most effectively. For each sensor, the noise voltage of the ODMR measurement is 0.057% of the signal voltage of the fluorescence.</p> Full article ">Figure 3
<p>Fabrication process of the NV centers in diamond using electron beam irradiation. A type Ib diamond after electron beam irradiation is divided into two and attached to each optical fiber.</p> Full article ">Figure 4
<p>(<b>a</b>) Gradiometer with the solenoid coil and the copper wire. The base length of the gradiometer was 27 mm. (<b>b</b>) Direction of the magnetic fields and the orientation axis of the NV center.</p> Full article ">Figure 5
<p>(<b>a</b>) Time-domain signals at sensor 1 (Ch.1), sensor 2 (Ch.2) and subtracted signal (Diff.). The detected wave for Ch. 2 is almost a single frequency wave because the copper wire applying a magnetic field of 30 Hz is far from sensor 2. On the other hand, since sensor 1 detects both 20 and 30 Hz magnetic fields, the detected wave at Ch.1 is a combination of the sin wave. (<b>b</b>) Fourier-transform of the time-domain signals from Ch.1 (solid red line), Ch. 2 (dashed-dotted green line) and diff (dashed blue line).</p> Full article ">Figure 6
<p>(<b>a</b>) Setup for measuring the base length dependence. Sensor 1 and sensor 2 are connected to optical fibers independently, allowing to change the base length freely. (<b>b</b>) Calculated differential signal level using the gradiometer as a function of the base length.</p> Full article ">Figure 7
<p>(<b>a</b>) Setup for measuring the magnetic noise frequency spectrum of the environment. (W/O difference, W/O shield) Only sensor 1 without any magnetic shielding nor using the gradiometer. (W/difference) The gradiometer with the base length of 50 mm. (W/ shield) Only sensor 1 in the three-layer magnetically shielded enclosure. (<b>b</b>) Three-layer magnetically shielded enclosure. The size of 70 cm is for magnetic sensing of small animals inside the enclosure. (<b>c</b>) Magnetic noise frequency spectrum. The peaks between 50 and 350 Hz are from the commercial power supply (50 Hz) and its harmonics. These magnetic noises propagated in space because the signal at 50 Hz and its harmonics disappear in a magnetically shielded enclosure (W/shield).</p> Full article ">Figure A1
<p>Energy-level diagram of the negatively charged NV center in diamond. The two magnetic sublevels with m<sub>S</sub> = ±1 split due to the Zeeman effect. The zero phonon line of the NV center is at 637 nm.</p> Full article ">
13 pages, 2094 KiB  
Communication
Development of an Automated Minimum Foot Clearance Measurement System: Proof of Principle
by Ghazaleh Delfi, Megan Kamachi and Tilak Dutta
Sensors 2021, 21(3), 976; https://doi.org/10.3390/s21030976 - 2 Feb 2021
Cited by 4 | Viewed by 2790
Abstract
Over half of older adult falls are caused by tripping. Many of these trips are likely due to obstacles present on walkways that put older adults or other individuals with low foot clearance at risk. Yet, Minimum Foot Clearance (MFC) values have not [...] Read more.
Over half of older adult falls are caused by tripping. Many of these trips are likely due to obstacles present on walkways that put older adults or other individuals with low foot clearance at risk. Yet, Minimum Foot Clearance (MFC) values have not been measured in real-world settings and existing methods make it difficult to do so. In this paper, we present the Minimum Foot Clearance Estimation (MFCE) system that includes a device for collecting calibrated video data from pedestrians on outdoor walkways and a computer vision algorithm for estimating MFC values for these individuals. This system is designed to be positioned at ground level next to a walkway to efficiently collect sagittal plane videos of many pedestrians’ feet, which is then processed offline to obtain MFC estimates. Five-hundred frames of video data collected from 50 different pedestrians was used to train (370 frames) and test (130 frames) a convolutional neural network. Finally, data from 10 pedestrians was analyzed manually by three raters and compared to the results of the network. The footwear detection network had an Intersection over Union of 85% and was able to find the bottom of a segmented shoe with a 3-pixel average error. Root Mean Squared (RMS) errors for the manual and automated methods for estimating MFC values were 2.32 mm, and 3.70 mm, respectively. Future work will compare the accuracy of the MFCE system to a gold standard motion capture system and the system will be used to estimate the distribution of MFC values for the population. Full article
(This article belongs to the Special Issue Artificial Intelligence and Internet of Things in Health Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>This image shows the Minimum Foot Clearance Estimation (MFCE) device. The device is made of a single video camera and two parallel laser pointers positioned on top of the camera. This device is meant to be placed on outdoor walkways to gather video of the feet and lower legs of pedestrians (general public) walking by. The camera is used to gather videos of passerby’s feet to assess their foot clearance. The parallel lasers have a fixed distance that acts as a reference scale in the recorded footage.</p> Full article ">Figure 2
<p>This image shows the MFCE device setup on the sidewalk. The device is placed on the street facing the sidewalk, with the center of the camera aligned to the sidewalk plane. The system will record passerby gait information and analyze the foot clearance automatically. No additional set-up is required.</p> Full article ">Figure 3
<p>This figure shows how much real-world distance (mm) is represented by each pixel for different camera-to-subject distances. Note that this is assuming that the camera has HD resolution (which was the case in our data collection). The accuracy of the measurements rely on how close the subject is to the camera. The farther away the subject, the more potential error in estimating the distances in the image.</p> Full article ">Figure 4
<p>This image shows an example of a frame where the laser dots are projected onto the pedestrian’s foot. The laser dots are marked with a green circle. The distance between the center of these two dots in the picture is equal to the fixed distance between the two laser pointers on the device.</p> Full article ">Figure 5
<p>This image provides an example labeled frame. The two instances of footwear are traced with a yellow line. The masks are then given to the network as the ground truth.</p> Full article ">Figure 6
<p>This figure provides the state diagram of the possible states of each frame in the captured pedestrian videos.</p> Full article ">Figure 7
<p>This figure shows the trajectory of the swing foot in one stride; the solid line shows the trajectory based on the lowest point on the detected swing masks and the dotted line shows the fitted 5th degree curve.</p> Full article ">Figure 8
<p>The Bland-Altman plot showing the inter-rater agreement between the five raters in the final evaluation.</p> Full article ">Figure 9
<p>The Manual measurements’ range within one standard deviation of each pedestrian’s respective mean is shown with the blue boxes and the automatic measurement for each case is marked by the green x.</p> Full article ">
12 pages, 4634 KiB  
Article
Changes of Corneal Biomechanical Properties upon Exclusive Ytt-/Sr-90 Irradiation of Pterygium
by Fritz Rigendinger, Daniel M. Aebersold, Zeljka Cvejic and Bojan Pajic
Sensors 2021, 21(3), 975; https://doi.org/10.3390/s21030975 - 2 Feb 2021
Cited by 1 | Viewed by 3037
Abstract
Background: It is known that pterygia above a certain size cause astigmatism and other aberrations of the human cornea and thus impair the quality of vision. Exclusive Sr-/Ytt-90 beta irradiation is a highly effective treatment for primary pterygia. The aim of this retrospective [...] Read more.
Background: It is known that pterygia above a certain size cause astigmatism and other aberrations of the human cornea and thus impair the quality of vision. Exclusive Sr-/Ytt-90 beta irradiation is a highly effective treatment for primary pterygia. The aim of this retrospective study is to determine the extent to which higher order corneal aberrations are affected by this treatment. Methods: Evaluation of corneal topographies and wavefront aberration data of 20 primary pterygia patients generated before and at different points in time in the first year after irradiation. Additionally, the size of the pterygium was measured. Results: The study showed a significant increase in coma and triple leaf aberrations in pterygia with a horizontal length of 2 mm and more. It was also found that a pterygium size greater than 2 mm significantly induces astigmatism. Both phenomena reduce visual quality. In none of the patients could a pterygium recurrence be detected after irradiation. Conclusions: If the pterygium size is less than 2 mm, early exclusive Sr/Ytt-90 beta irradiation can be recommended. If the size is more than 2 mm, a pterygium excision 6 months after beta irradiation can be discussed. Full article
(This article belongs to the Special Issue Emerging Technologies for Sensing, Imaging and Biomechanics of the Eye for Diagnostics and Surgery)
►▼ Show Figures

Figure 1

Figure 1
<p>Islets of corneal epithelial cells arranged within the fibrous structures of the pterygium after beta irradiation.</p> Full article ">Figure 2
<p>Confocal microscope images of a pterygium within the surface at 0 µm, through layers in deeper positions at 80 and 100 µm to the sclera at 120 µm.</p> Full article ">Figure 3
<p>Stronium-/Yttrium-90 beta-irradiation decay scheme.</p> Full article ">Figure 4
<p>Topographical, schematic illustration shows the location on the nasal corneal limbus where the radiation applicator was placed. Note the cornea flattens in the area of the pterygium.</p> Full article ">Figure 5
<p>The diagram shows the penetration depth of beta radiation in the tissue. It can be seen that only 6% of the irradiation penetrates 4 mm, i.e., to the equator of the lens.</p> Full article ">Figure 6
<p>A Pentacam–Scheimpflug image (inverted colours) shows a nasal pterygium in form of corneal thickening (arrow) of the left eye of a 62-year-old male with a horizontal length of 3.5 mm before irradiation. By merging 25 Scheimpflug images, the 3D model of the anterior eye segment and all other calculations are generated.</p> Full article ">Figure 7
<p>A typical Pentacam display of the same nasal pterygium shown in <a href="#sensors-21-00975-f005" class="html-fig">Figure 5</a>. Clockwise starting from top left: anterior sagittal curvature map, anterior elevation map, posterior elevation map, corneal thickness map (pachymetry). The pterygium thickens and flattens the peripheral nasal cornea and thus reduces local refractive power.</p> Full article ">Figure 8
<p>Zernike polynomials as displayed by the Pentacam software. The red framed aberrations increased significantly after exclusive Strontium-/Yttrium-90 beta irradiation of primary pterygia. The majority of HOAs (higher order aberrations) of the 3rd order, but still the overall RMS (root mean square) for HOAs of the 3rd to 6th order showed no statistical significance.</p> Full article ">Figure 9
<p>Shown is the regression of the pterygium after Sr-/Ytt-90 beta irradiation. The red arrows indicate where the border of the pterygium was before irradiation and the yellow arrows mark the border 12 months after irradiation.</p> Full article ">
24 pages, 10548 KiB  
Article
Autonomous Vision-Based Primary Distribution Systems Porcelain Insulators Inspection Using UAVs
by Ehab Ur Rahman, Yihong Zhang, Sohail Ahmad, Hafiz Ishfaq Ahmad and Sayed Jobaer
Sensors 2021, 21(3), 974; https://doi.org/10.3390/s21030974 - 2 Feb 2021
Cited by 52 | Viewed by 5600
Abstract
The early detection of damaged (partially broken) outdoor insulators in primary distribution systems is of paramount importance for continuous electricity supply and public safety. Unmanned aerial vehicles (UAVs) present a safer, autonomous, and efficient way to examine the power system components without closing [...] Read more.
The early detection of damaged (partially broken) outdoor insulators in primary distribution systems is of paramount importance for continuous electricity supply and public safety. Unmanned aerial vehicles (UAVs) present a safer, autonomous, and efficient way to examine the power system components without closing the power distribution system. In this work, a novel dataset is designed by capturing real images using UAVs and manually generated images collected to overcome the data insufficiency problem. A deep Laplacian pyramid-based super-resolution network is implemented to reconstruct high-resolution training images. To improve the visibility of low-light images, a low-light image enhancement technique is used for the robust exposure correction of the training images. A different fine-tuning strategy is implemented for fine-tuning the object detection model to increase detection accuracy for the specific faulty insulators. Several flight path strategies are proposed to overcome the shuttering effect of insulators, along with providing a less complex and time- and energy-efficient approach for capturing a video stream of the power system components. The performance of different object detection models is presented for selecting the most suitable one for fine-tuning on the specific faulty insulator dataset. For the detection of damaged insulators, our proposed method achieved an F1-score of 0.81 and 0.77 on two different datasets and presents a simple and more efficient flight strategy. Our approach is based on real aerial inspection of in-service porcelain insulators by extensive evaluation of several video sequences showing robust fault recognition and diagnostic capabilities. Our approach is demonstrated on data acquired by a drone in Swat, Pakistan. Full article
(This article belongs to the Section Electronic Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Depiction of insulators in overhead power distribution lines.</p> Full article ">Figure 2
<p>Illustration of overhead power distribution line from drone camera.</p> Full article ">Figure 3
<p>Workflow for the proposed methodology.</p> Full article ">Figure 4
<p>Power system components mounted on a supporting electric pole.</p> Full article ">Figure 5
<p>Different faults found in power distribution systems: (<b>a</b>) broken suspension disc insulator, (<b>b</b>) broken cross arm, (<b>c</b>) unclipped pin insulator from the cross arm, (<b>d</b>) broken pin insulator.</p> Full article ">Figure 6
<p>Internal structure of pin and suspension disc porcelain insulators: (<b>a</b>) internal structure of a suspension disc insulator, (<b>b</b>) the length parameters of the pin insulators.</p> Full article ">Figure 7
<p>Normal (intact) and defective (broken) porcelain insulators with different view angles.</p> Full article ">Figure 8
<p>Steps of dataset formation from (<b>a</b>–<b>e</b>).</p> Full article ">Figure 9
<p>Components of the UAV and the prototype of the electric pole with broken insulators.</p> Full article ">Figure 10
<p>(<b>a</b>) Workflow of LapSRN for upsampling a low-resolution image, (<b>b</b>) LR patch extraction.</p> Full article ">Figure 11
<p>Low-light enhancement using LIME on different training images.</p> Full article ">Figure 12
<p>Fine-tuning strategy for insulator detection model.</p> Full article ">Figure 13
<p>Flightpath strategies for the UAV for inspection of the insulators mounted on an electric pole. (<b>a</b>) Presenting a circular flight strategy (<b>b</b>) Presenting a straight flight path strategy.</p> Full article ">Figure 14
<p>Aerial image of 3-phase 11kV lines showing the spacing between the conductor lines and obstacle-free path for the drone flight.</p> Full article ">Figure 15
<p>Improvement steps of specific insulator training dataset by LapSRN and LIME.</p> Full article ">Figure 16
<p>Precision recall curve of YoloV4 for normal insulator detection model for different test datasets.</p> Full article ">Figure 17
<p>Precision vs. recall curve for the faulty insulator detection model with the proposed methodology.</p> Full article ">Figure A1
<p>YoloV4 architecture with different blocks [<a href="#B42-sensors-21-00974" class="html-bibr">42</a>].</p> Full article ">Figure A2
<p>(<b>a</b>) Shuttering due to conductor and clipping wire. (<b>b</b>) Faulty insulator training image. (<b>c</b>) Normal insulator training image.</p> Full article ">Figure A3
<p>Test dataset (<b>a</b>), complex (C) (<b>b</b>), low light (R-1) (<b>c</b>), aerial image with better visibility (<b>d</b>), Prototype-C (<b>e</b>), Prototype-S, (<b>f</b>) variants of a low-light image with LIME.</p> Full article ">Figure A4
<p>Test results. Row 1, left: YoloV4. Row 1, right: YoloV5x. Row 2, left: YoloV4 Tiny. Row 2, right: YoloV5s.</p> Full article ">Figure A5
<p>Test results of the proposed YoloV4-based faulty insulator detection model with flight path patterns.</p> Full article ">
19 pages, 2586 KiB  
Article
A Spectral-Based Approach for BCG Signal Content Classification
by Mohamed Chiheb Ben Nasr, Sofia Ben Jebara, Samuel Otis, Bessam Abdulrazak and Neila Mezghani
Sensors 2021, 21(3), 1020; https://doi.org/10.3390/s21031020 - 2 Feb 2021
Cited by 7 | Viewed by 4271
Abstract
This paper has two objectives: the first is to generate two binary flags to indicate useful frames permitting the measurement of cardiac and respiratory rates from Ballistocardiogram (BCG) signals—in fact, human body activities during measurements can disturb the BCG signal content, leading to [...] Read more.
This paper has two objectives: the first is to generate two binary flags to indicate useful frames permitting the measurement of cardiac and respiratory rates from Ballistocardiogram (BCG) signals—in fact, human body activities during measurements can disturb the BCG signal content, leading to difficulties in vital sign measurement; the second objective is to achieve refined BCG signal segmentation according to these activities. The proposed framework makes use of two approaches: an unsupervised classification based on the Gaussian Mixture Model (GMM) and a supervised classification based on K-Nearest Neighbors (KNN). Both of these approaches consider two spectral features, namely the Spectral Flatness Measure (SFM) and Spectral Centroid (SC), determined during the feature extraction step. Unsupervised classification is used to explore the content of the BCG signals, justifying the existence of different classes and permitting the definition of useful hyper-parameters for effective segmentation. In contrast, the considered supervised classification approach aims to determine if the BCG signal content allows the measurement of the heart rate (HR) and the respiratory rate (RR) or not. Furthermore, two levels of supervised classification are used to classify human-body activities into many realistic classes from the BCG signal (e.g., coughing, holding breath, air expiration, movement, et al.). The first one considers frame-by-frame classification, while the second one, aiming to boost the segmentation performance, transforms the frame-by-frame SFM and SC features into temporal series which track the temporal variation of the measures of the BCG signal. The proposed approach constitutes a novelty in this field and represents a powerful method to segment BCG signals according to human body activities, resulting in an accuracy of 94.6%. Full article
(This article belongs to the Special Issue Selected Papers from International Conference on Smart Living and Public Health (ICOST 2020))
►▼ Show Figures

Figure 1

Figure 1
<p>Flowchart of adopted methodology. BCG: Ballistocardiogram. CAD: cardiac activity detection; RAD: respiratory activity detection.</p> Full article ">Figure 2
<p>Illustration of the experimental protocol activities [<a href="#B16-sensors-21-01020" class="html-bibr">16</a>].</p> Full article ">Figure 3
<p>Illustration of the BCG signal during the experimental protocol activities. Each color corresponds to a human body activity.</p> Full article ">Figure 4
<p>Theoretical respiratory and cardiac activities present in the BCG waveform.</p> Full article ">Figure 5
<p>Microbend FOS principle. The light passing through the microbend FOS is modulated by the deformations in the optical fiber due to the displacement of the micro-benders [<a href="#B23-sensors-21-01020" class="html-bibr">23</a>].</p> Full article ">Figure 6
<p>Steps of time-series generation.</p> Full article ">Figure 7
<p>Illustration of temporal evolution and frequency representation of BCG signals. (<b>a</b>) normal respiration in still position, (<b>b</b>) during coughing activity, (<b>c</b>) while holding breath.</p> Full article ">Figure 8
<p>Distribution of the Spectral Flatness Measure (SFM) and Spectral Centroid (SC) values in the dataset.</p> Full article ">Figure 9
<p>Histograms of SFM and SC values in the dataset.</p> Full article ">Figure 10
<p>Representation of the Akaike Information Criterion (<math display="inline"><semantics> <mrow> <mi>A</mi> <mi>I</mi> <mi>C</mi> </mrow> </semantics></math>) and Bayes Information Criterion (<math display="inline"><semantics> <mrow> <mi>B</mi> <mi>I</mi> <mi>C</mi> </mrow> </semantics></math>) as a function of the number of Gaussian components.</p> Full article ">Figure 11
<p>Variation of the mean of silhouette over the frames length.</p> Full article ">
20 pages, 1782 KiB  
Article
Intelligent Decision-Making of Scheduling for Dynamic Permutation Flowshop via Deep Reinforcement Learning
by Shengluo Yang, Zhigang Xu and Junyi Wang
Sensors 2021, 21(3), 1019; https://doi.org/10.3390/s21031019 - 2 Feb 2021
Cited by 41 | Viewed by 5645
Abstract
Dynamic scheduling problems have been receiving increasing attention in recent years due to their practical implications. To realize real-time and the intelligent decision-making of dynamic scheduling, we studied dynamic permutation flowshop scheduling problem (PFSP) with new job arrival using deep reinforcement learning (DRL). [...] Read more.
Dynamic scheduling problems have been receiving increasing attention in recent years due to their practical implications. To realize real-time and the intelligent decision-making of dynamic scheduling, we studied dynamic permutation flowshop scheduling problem (PFSP) with new job arrival using deep reinforcement learning (DRL). A system architecture for solving dynamic PFSP using DRL is proposed, and the mathematical model to minimize total tardiness cost is established. Additionally, the intelligent scheduling system based on DRL is modeled, with state features, actions, and reward designed. Moreover, the advantage actor-critic (A2C) algorithm is adapted to train the scheduling agent. The learning curve indicates that the scheduling agent learned to generate better solutions efficiently during training. Extensive experiments are carried out to compare the A2C-based scheduling agent with every single action, other DRL algorithms, and meta-heuristics. The results show the well performance of the A2C-based scheduling agent considering solution quality, CPU times, and generalization. Notably, the trained agent generates a scheduling action only in 2.16 ms on average, which is almost instantaneous and can be used for real-time scheduling. Our work can help to build a self-learning, real-time optimizing, and intelligent decision-making scheduling system. Full article
(This article belongs to the Special Issue Artificial Intelligence for Smart Manufacturing: Methods and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>The system architecture of solving dynamic PFSP using DRL.</p> Full article ">Figure 2
<p>Average total tardiness cost on all test instances at each training epoch.</p> Full article ">Figure 3
<p>Average episode reward on all test instances at each training epoch.</p> Full article ">Figure 4
<p>Comparison between A2C and SDR considering average total tardiness cost on all test instances.</p> Full article ">Figure 5
<p>Average episode cost on all test instances at each training epoch for DQN and DDQN.</p> Full article ">Figure 6
<p>Average episode reward on all test instances at each training epoch for DQN and DDQN.</p> Full article ">Figure 7
<p>Average total tardiness cost of compared algorithms on all test instances. Three replications are carried out for each test instance. IG and GA are tested under two search iteration levels, 50 iterations and 300 iterations.</p> Full article ">Figure 8
<p>Comparison between A2C and SDR considering average total tardiness cost on extended instances.</p> Full article ">Figure 9
<p>The percentages that A2C outperforms SDR considering average total tardiness cost on the original and extended instances.6. Conclusions.</p> Full article ">
19 pages, 43772 KiB  
Article
Emotion Recognition Based on Skin Potential Signals with a Portable Wireless Device
by Shuhao Chen, Ke Jiang, Haoji Hu, Haoze Kuang, Jianyi Yang, Jikui Luo, Xinhua Chen and Yubo Li
Sensors 2021, 21(3), 1018; https://doi.org/10.3390/s21031018 - 2 Feb 2021
Cited by 13 | Viewed by 4825
Abstract
Emotion recognition is of great importance for artificial intelligence, robots, and medicine etc. Although many techniques have been developed for emotion recognition, with certain successes, they rely heavily on complicated and expensive equipment. Skin potential (SP) has been recognized to be correlated with [...] Read more.
Emotion recognition is of great importance for artificial intelligence, robots, and medicine etc. Although many techniques have been developed for emotion recognition, with certain successes, they rely heavily on complicated and expensive equipment. Skin potential (SP) has been recognized to be correlated with human emotions for a long time, but has been largely ignored due to the lack of systematic research. In this paper, we propose a single SP-signal-based method for emotion recognition. Firstly, we developed a portable wireless device to measure the SP signal between the middle finger and left wrist. Then, a video induction experiment was designed to stimulate four kinds of typical emotion (happiness, sadness, anger, fear) in 26 subjects. Based on the device and video induction, we obtained a dataset consisting of 397 emotion samples. We extracted 29 features from each of the emotion samples and used eight well-established algorithms to classify the four emotions based on these features. Experimental results show that the gradient-boosting decision tree (GBDT), logistic regression (LR) and random forest (RF) algorithms achieved the highest accuracy of 75%. The obtained accuracy is similar to, or even better than, that of other methods using multiple physiological signals. Our research demonstrates the feasibility of the SP signal’s integration into existing physiological signals for emotion recognition. Full article
(This article belongs to the Special Issue Sensor-Based Human Activity Monitoring)
►▼ Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Block diagram of the circuit functions of the portable device. (<b>b</b>) The portable device for skin potential (SP) signal acquisition. (<b>c</b>) The screenshot of the application when it receives signals.</p> Full article ">Figure 2
<p>Characteristics the of SP signal. (<b>a</b>) The original signal spectrum, and (<b>b</b>) frequency spectrum analysis of the signal.</p> Full article ">Figure 3
<p>(<b>a</b>) SP record diagram of six points from a subject during video viewing. (<b>b</b>) The testing points.</p> Full article ">Figure 4
<p>(<b>a</b>) The experimental scene and a screenshot of the SP signal on the mobile phone; (<b>b</b>) The scene of happiness from the first part of the video (excerpt from the movie NEVER SAY DIE); (<b>c</b>) The scene of sadness from the second part of the video (excerpt from the movie Aftershock); (<b>d</b>) The scene of anger from the third part of the video (excerpt from the movie God Of Gamblers); (<b>e</b>) The scene of fear from the fourth part of the video (excerpt from the movie Insidious); (<b>f</b>) The landscape image shown between two adjacent video parts.</p> Full article ">Figure 5
<p>Block diagram of experimental procedures.</p> Full article ">Figure 6
<p>Block diagram of the construction procedure of SP-based emotion recognition model.</p> Full article ">Figure 7
<p>The whole SP recording from subject 5 during video viewing after data preprocessing.</p> Full article ">Figure 8
<p>Twelve emotion samples of happiness, sadness, anger and fear.</p> Full article ">Figure 9
<p>The classification accuracy on the test set of eight classifiers when (<b>a</b>) 15 features are selected; (<b>b</b>) 20 features are selected; (<b>c</b>) 25 features are selected; (<b>d</b>) 29 features are selected.</p> Full article ">Figure 10
<p>The contribution rates of all 29 features.</p> Full article ">
10 pages, 6068 KiB  
Communication
Software for Matching Standard Activity Enzyme Biosensors for Soil Pollution Analysis
by Valentina A. Kratasyuk, Elizaveta M. Kolosova, Oleg S. Sutormin, Viktoriya I. Lonshakova-Mukina, Matvey M. Baygin, Nadezhda V. Rimatskaya, Irina E. Sukovataya and Alexander A. Shpedt
Sensors 2021, 21(3), 1017; https://doi.org/10.3390/s21031017 - 2 Feb 2021
Cited by 7 | Viewed by 3794
Abstract
This work is dedicated to developing enzyme biosensor software to solve problems regarding soil pollution analysis. An algorithm and specialised software have been developed which stores, analyses and visualises data using JavaScript programming language. The developed software is based on matching data of [...] Read more.
This work is dedicated to developing enzyme biosensor software to solve problems regarding soil pollution analysis. An algorithm and specialised software have been developed which stores, analyses and visualises data using JavaScript programming language. The developed software is based on matching data of 51 non-commercial standard soil samples and their inhibitory effects on three enzyme systems of varying complexity. This approach is able to identify the influence of chemical properties soil samples, without toxic agents, on enzyme biosensors. Such software may find wide use in environmental monitoring. Full article
(This article belongs to the Special Issue Fluorescence and Chemical Luminescence Sensors)
►▼ Show Figures

Figure 1

Figure 1
<p>Standard soils: (<b>a</b>) view of vessels containing basic soil substrates, (<b>b</b>) the experiment scheme based on the basic soil substrates.</p> Full article ">Figure 2
<p>Flowchart of the search of the reference with a minimum deviation using Euclidean Distance.</p> Full article ">Figure 3
<p>View of the application: (<b>a</b>) when launched in a browser; (<b>b</b>) when completing the input boxes and selecting a suitable standard soil.</p> Full article ">Figure 4
<p>Application graph, the “reference” category is not active (crossed out) and the data is only displayed for “sample”.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-347
Previous Issue
Next Issue
Back to TopTop