Nothing Special   »   [go: up one dir, main page]

You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 

Perception and Detection of Intelligent Vision

Special Issue Editors


E-Mail Website
Guest Editor
School of Robotics, Hunan University, Changsha 410082, China.
Interests: robotic perception; machine learning; pattern recognition

E-Mail Website
Guest Editor
School of Engineering and Design, Hunan Normal University, Changsha 410081, China
Interests: computer vision; deep learning; few-shot learning; representation learning

E-Mail Website
Guest Editor
The School of Artificial Intelligence, Xidian University, Xi'an 710126, China
Interests: computer vision; 3D vision; scene understanding
Special Issues, Collections and Topics in MDPI journals

E-Mail Website
Guest Editor
Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China
Interests: intelligent robots and perception; action recognition and human-computer interaction; few-shot learning

Special Issue Information

Dear Colleagues,

Vision plays the most important role in the cognitive system of humans. Constructing an intelligent vision system that can achieve part or even all of the capabilities of the human vision system is one of the ultimate goals of many research areas such as computer vison, artificial intelligence, and cognitive science.  In recent years, based on the significant amount of internet data, data-driven methods have greatly advanced the frontiers of intelligent vision, although it is not yet as powerful as the human vision system.

The aim of this Special Issue is to explore recent advances in visual perception in relation to the fields of computer vision and cognitive science. This Special Issue will bring together leading researchers and developers to present their latest research on algorithm design, system frameworks, and cognitive theories for developing intelligent vision systems. In this Special Issue, original research articles and reviews are welcome. Research areas may include (but not are limited to) the following:

  • Computer vision;
  • Robot vision;
  • Visual perception;
  • Scene understanding;
  • 3D vision;
  • Deep learning;
  • Visual representation learning;
  • Intelligent vision device;
  • Large vision model;
  • Unsupervised learning;
  • Multi-model learning;
  • Visual cognition theories;
  • Action recognition and understanding;
  • Human-computer interaction. 

We look forward to receiving your contributions. 

Prof. Dr.  Hongshan Yu
Dr. Zhengeng Yang
Dr. Mingtao Feng
Dr. Qieshi Zhang
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Big Data and Cognitive Computing is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • visual perception
  • deep learning
  • intelligent vision systems
  • vision cognition
  • unsupervised learning
  • large models

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue polices can be found here.

Published Papers (3 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

18 pages, 3292 KiB  
Article
Suspension Parameter Estimation Method for Heavy-Duty Freight Trains Based on Deep Learning
by Changfan Zhang, Yuxuan Wang and Jing He
Big Data Cogn. Comput. 2024, 8(12), 181; https://doi.org/10.3390/bdcc8120181 - 4 Dec 2024
Abstract
The suspension parameters of heavy-duty freight trains can deviate from their initial design values due to material aging and performance degradation. While traditional multibody dynamics simulation models are usually designed for fixed working conditions, it is difficult for them to adequately analyze the [...] Read more.
The suspension parameters of heavy-duty freight trains can deviate from their initial design values due to material aging and performance degradation. While traditional multibody dynamics simulation models are usually designed for fixed working conditions, it is difficult for them to adequately analyze the safety status of the vehicle–line system in actual operation. To address this issue, this research provides a suspension parameter estimation technique based on CNN-GRU. Firstly, a prototype C80 train was utilized to build a simulation model for multibody dynamics. Secondly, six key suspension parameters for wheel–rail force were selected using the Sobol global sensitivity analysis method. Then, a CNN-GRU proxy model was constructed, with the actually measured wheel–rail forces as a reference. By combining this approach with NSGA-II (Non-dominated Sorting Genetic Algorithm II), the key suspension parameters were calculated. Finally, the estimated parameter values were applied into the vehicle–line coupled multibody dynamical model and validated. The results show that, with the corrected dynamical model, the relative errors of the simulated wheel–rail force are reduced from 9.28%, 6.24% and 18.11% to 7%, 4.52% and 10.44%, corresponding to straight, curve, and long and steep uphill conditions, respectively. The wheel–rail force simulation’s precision is increased, indicating that the proposed method is effective in estimating the suspension parameters for heavy-duty freight trains. Full article
(This article belongs to the Special Issue Perception and Detection of Intelligent Vision)
14 pages, 7338 KiB  
Article
Strawberry Ripeness Detection Using Deep Learning Models
by Zhiyuan Mi and Wei Qi Yan
Big Data Cogn. Comput. 2024, 8(8), 92; https://doi.org/10.3390/bdcc8080092 - 15 Aug 2024
Cited by 3 | Viewed by 1760
Abstract
In agriculture, the timely and accurate assessment of fruit ripeness is crucial to optimizing harvest planning and reduce waste. In this article, we explore the integration of two cutting-edge deep learning models, YOLOv9 and Swin Transformer, to develop a complex model for detecting [...] Read more.
In agriculture, the timely and accurate assessment of fruit ripeness is crucial to optimizing harvest planning and reduce waste. In this article, we explore the integration of two cutting-edge deep learning models, YOLOv9 and Swin Transformer, to develop a complex model for detecting strawberry ripeness. Trained and tested on a specially curated dataset, our model achieves a mean precision (mAP) of 87.3% by using the metric intersection over union (IoU) at a threshold of 0.5. This outperforms the model using YOLOv9 alone, which achieves an mAP of 86.1%. Our model also demonstrated improved precision and recall, with precision rising to 85.3% and recall rising to 84.0%, reflecting its ability to accurately and consistently detect different stages of strawberry ripeness. Full article
(This article belongs to the Special Issue Perception and Detection of Intelligent Vision)
Show Figures

Figure 1

Figure 1
<p>Overall structure of strawberry ripeness detection model.</p>
Full article ">Figure 2
<p>YOLOv9 structure.</p>
Full article ">Figure 3
<p>Swin Transformer structure.</p>
Full article ">Figure 4
<p>Swin Transformer blocks.</p>
Full article ">Figure 5
<p>Patch Merging.</p>
Full article ">Figure 6
<p>MSA and W-MSA.</p>
Full article ">Figure 7
<p>Samples of our dataset.</p>
Full article ">Figure 8
<p>Example of results after labeling.</p>
Full article ">Figure 9
<p>Data splitting pie chart.</p>
Full article ">Figure 10
<p>An example of bounding boxes. Red boxes are predicted regions, green boxes are ground truth.</p>
Full article ">Figure 11
<p>Results of our model on validation set.</p>
Full article ">Figure 12
<p>PR curve of YOLOv9+Swin Transformer model.</p>
Full article ">Figure 13
<p>PR curve of YOLOv9 model.</p>
Full article ">Figure 14
<p>Plots of results of YOLOv9+Swin Transformer model.</p>
Full article ">
24 pages, 1853 KiB  
Article
Optimal Image Characterization for In-Bed Posture Classification by Using SVM Algorithm
by Claudia Angelica Rivera-Romero, Jorge Ulises Munoz-Minjares, Carlos Lastre-Dominguez and Misael Lopez-Ramirez
Big Data Cogn. Comput. 2024, 8(2), 13; https://doi.org/10.3390/bdcc8020013 - 26 Jan 2024
Cited by 3 | Viewed by 2156
Abstract
Identifying patient posture while they are lying in bed is an important task in medical applications such as monitoring a patient after a surgical intervention, sleep supervision to identify behavioral and physiological markers, or for bedsore prevention. An acceptable strategy to identify the [...] Read more.
Identifying patient posture while they are lying in bed is an important task in medical applications such as monitoring a patient after a surgical intervention, sleep supervision to identify behavioral and physiological markers, or for bedsore prevention. An acceptable strategy to identify the patient’s position is the classification of images created from a grid of pressure sensors located in the bed. These samples can be arranged based on supervised learning methods. Usually, image conditioning is required before images are loaded into a learning method to increase classification accuracy. However, continuous monitoring of a person requires large amounts of time and computational resources if complex pre-processing algorithms are used. So, the problem is to classify the image posture of patients with different weights, heights, and positions by using minimal sample conditioning for a specific supervised learning method. In this work, it is proposed to identify the patient posture from pressure sensor images by using well-known and simple conditioning techniques and selecting the optimal texture descriptors for the Support Vector Machine (SVM) method. This is in order to obtain the best classification and to avoid image over-processing in the conditioning stage for the SVM. The experimental stages are performed with the color models Red, Green, and Blue (RGB) and Hue, Saturation, and Value (HSV). The results show an increase in accuracy from 86.9% to 92.9% and in kappa value from 0.825 to 0.904 using image conditioning with histogram equalization and a median filter, respectively. Full article
(This article belongs to the Special Issue Perception and Detection of Intelligent Vision)
Show Figures

Figure 1

Figure 1
<p>Postures of patients are considered for the identification of the pressure position: (<b>a</b>) dorsal decubitus, (<b>b</b>) lateral decubitus, (<b>c</b>) lateral decubitus with an external object, and (<b>d</b>) dorsal decubitus with crossed legs.</p>
Full article ">Figure 2
<p>Methodology for patient posture identification.</p>
Full article ">Figure 3
<p>ANOVA of the features: (<b>a</b>) mean values of CPROM<sub><span class="html-italic">S</span></sub>, (<b>b</b>) mean values of DVARH<sub><span class="html-italic">G</span></sub>. The red plus sign (+) in both figures indicates the outliers.</p>
Full article ">Figure 4
<p>Features maps: (<b>a</b>) feature CONTR<sub><span class="html-italic">H</span></sub> versus SAVGH<sub><span class="html-italic">B</span></sub> of <math display="inline"><semantics> <msub> <mi>P</mi> <mn>1</mn> </msub> </semantics></math> and <math display="inline"><semantics> <msub> <mi>P</mi> <mn>4</mn> </msub> </semantics></math>; and (<b>b</b>) feature CORRM<sub><span class="html-italic">R</span></sub> versus DVARH<sub><span class="html-italic">G</span></sub> of <math display="inline"><semantics> <msub> <mi>P</mi> <mn>3</mn> </msub> </semantics></math> and <math display="inline"><semantics> <msub> <mi>P</mi> <mn>4</mn> </msub> </semantics></math>.</p>
Full article ">Figure 5
<p>Training data in 2D and 3D hyperplanes, circles and asterisks are the class data of position <span class="html-italic">P</span><sub>1</sub> and <span class="html-italic">P</span><sub>2</sub>, respectively. (<b>a</b>) features SOSVH<sub><span class="html-italic">S</span></sub> and IDMNC<sub><span class="html-italic">V</span></sub> for data in <math display="inline"><semantics> <msub> <mi>P</mi> <mn>1</mn> </msub> </semantics></math>, (<b>b</b>) features SOSVH<sub><span class="html-italic">S</span></sub> and IDMNC<sub><span class="html-italic">V</span></sub> for data in <math display="inline"><semantics> <msub> <mi>P</mi> <mn>2</mn> </msub> </semantics></math>, and (<b>c</b>) 2D hyperplane for both classes.</p>
Full article ">Figure 6
<p>Confusion matrices obtained for the multiple classifications for characterized images in set <math display="inline"><semantics> <msub> <mi>S</mi> <mn>1</mn> </msub> </semantics></math>: (<b>a</b>) classified data in <math display="inline"><semantics> <msub> <mi>V</mi> <mn>1</mn> </msub> </semantics></math> samples in color components of RGB, and (<b>b</b>) classified data in <math display="inline"><semantics> <msub> <mi>V</mi> <mn>2</mn> </msub> </semantics></math> samples in color components of HSV.</p>
Full article ">Figure 7
<p>Confusion matrices obtained for the multiple classifications: (<b>a</b>) classified data with PCA components in <math display="inline"><semantics> <msub> <mi>V</mi> <mn>1</mn> </msub> </semantics></math> samples in color components of RGB, and (<b>b</b>) classified data with PCA components in <math display="inline"><semantics> <msub> <mi>V</mi> <mn>2</mn> </msub> </semantics></math> samples in color components of HSV.</p>
Full article ">Figure 8
<p>ROC curves obtained for the multiple classifications: (<b>a</b>) classified data in <math display="inline"><semantics> <msub> <mi>V</mi> <mn>1</mn> </msub> </semantics></math> samples in color components of RGB, and (<b>b</b>) classified data in <math display="inline"><semantics> <msub> <mi>V</mi> <mn>2</mn> </msub> </semantics></math> samples in color components of HSV.</p>
Full article ">Figure 9
<p>ROC curves obtained for the multiple classifications: (<b>a</b>) classified data with PCA components in <math display="inline"><semantics> <msub> <mi>V</mi> <mrow> <mi>R</mi> <mi>G</mi> <mi>B</mi> </mrow> </msub> </semantics></math> samples in color components of RGB, and (<b>b</b>) classified data with PCA components in <math display="inline"><semantics> <msub> <mi>V</mi> <mrow> <mi>H</mi> <mi>S</mi> <mi>V</mi> </mrow> </msub> </semantics></math> samples in color components of HSV.</p>
Full article ">
Back to TopTop