Search Results (10,399)

In recent years, anchor-free object detectors have become predominant in deep learning, the YOLOv8 model as a real-time object detector based on anchor-free frames is universal and influential, it efficiently detects objects across multiple scales. However, the generalization performance of the model is lacking, and the feature fusion within the neck module overly relies on its structural design and dataset size, and it is particularly difficult to localize and detect small objects. To address these issues, we propose the FL-YOLOv8 object detector, which is improved based on YOLOv8s. Firstly, we introduce the FSDI module in the neck, enhancing semantic information across all layers and incorporating rich detailed features through straightforward layer-hopping connections. This module integrates both high-level and low-level information to enhance the accuracy and efficiency of image detection. Meanwhile, the structure of the model was optimized and designed, and the LSCD module is constructed in the detection head; adopting a lightweight shared convolutional detection head reduces the number of parameters and computation of the model by 19% and 10%, respectively. Our model achieves a comprehensive performance of 45.5% on the COCO generalized dataset, surpassing the benchmark by 0.8 percentage points. To further validate the effectiveness of the method, experiments were also performed on specific domain urine sediment data (FCUS22), and the results on category detection also better justify the FL-YOLOv8 object detection algorithm. Full article

Oral histories of Latina domestic workers in the United States feature hybrid narratives combining accounts of illness and “toxic discourse”. We approach domestic workers’ illnesses and disabilities in a capacious, extra-medical context that registers multiple axes of precarity (economic, racial, and migratory). We are naming this context “settler maintenance”. Riffing on the specific and general valences of “maintenance” (i.e., as a synonym for cleaning work, and as a term for the practices and ideologies involved in a structure’s upkeep), this term has multiple meanings. First, it describes U.S. domestic workers’ often-compulsory use of hazardous chemical agents that promise to remove dirt speedily, yet that imperil domestic workers’ health. The use of these chemicals perpetuates two other, more abstract kinds of settler maintenance: (1) the continuation of socioeconomic hierarchies between immigrant domestic workers and settler employers, and (2) the continuation of (white) settlers’ extractive relationship to the land qua private property. To challenge this logic of settler maintenance, which is predicated on a lack of care for care workers, Latina domestic workers have developed alternative forms of care via lateral networks and political activism. Full article

To tackle the issue of limited sample data in small sample fault diagnosis for rolling bearings using deep learning, we propose a fault diagnosis method that integrates a KANs-CNN network. Initially, the raw vibration signals are converted into two-dimensional time-frequency images via a continuous wavelet transform. Next, Using CNN combined with KANs for feature extraction, the nonlinear activation of KANs helps extract deep and complex features from the data. After the output of CNN-KANs, an FAN network module is added. The FAN module can employ various feature aggregation strategies, such as weighted averaging, max pooling, addition aggregation, etc., to combine information from multiple feature levels. To further tackle the small sample issue, data generation is performed on the original data through diffusion networks under conditions of fewer samples for bearings and tools, thereby increasing the sample size of the dataset and enhancing fault diagnosis accuracy. Experimental results demonstrate that, under small sample conditions, this method achieves higher accuracy compared to other approaches. Full article

Highly accurate rice cultivation distribution and area extraction are essential to food security. Moreover, Inner Mongolia, whose slogan is “from scientific rice to world rice”, is an essential national rice production base. However, high-quality rice mapping products at high resolutions are still scarce around the Inner Mongolia Autonomous Region. This condition is not conducive to rational planning of farmland resources, maintaining food security, and promoting sustainable growth of the local agricultural economy. In this study, the rice backscattering intensity difference index from the vertically polarized backscatter intensity of Sentinel-1 and the phenology differential index from the spectral indices of two critical rice phenological phases of Sentinel-2 images were constructed. Other spectral features, including spectral indices, tasseled cap, and texture features, were computed using simple non-iterative clustering (SNIC) to achieve image segmentation. These variables served as input features for the random forest (RF) algorithm. Results reveal that employing the RF with the SNIC segmentation algorithm and combining it with optical and synthetic aperture radar data is an effective way to extract data on rice in mid-latitude regions. The overall accuracy and kappa coefficient are 0.98 and 0.967, correspondingly. The accuracy for rice is 0.99, as proven by empirical data. These results meet the requirements of regional rice cultivation assessment and area monitoring. Furthermore, owing to its resilience against longitude-associated influences, the model discerns rice across diverse regions and multiple years, achieving an R2 of 0.99. This capability significantly bolsters efforts to improve regional food security and the pursuit of sustainable development. Full article

Background and Objectives: The link between celiac disease (CD) and type 1 diabetes (T1D) has been well-documented in the medical literature and is thought to be due to a shared genetic predisposition in addition to environmental triggers. This study aimed to determine the seroprevalence and biopsy-proven CD (PBCD) prevalence in individuals with T1D from Saudi Arabia and identify their clinical characteristics and the impact on glycemic control. Materials and Methods: A total of 969 children and adolescents with confirmed T1D were investigated. Prospective and retrospective data were collected to include clinical, anthropometric, and biochemical data. Total IgA and anti-TTG-IgA antibodies were screened to detect seropositive cases. Upper intestinal endoscopy and biopsy were performed to find BPCD. Results: The seroprevalence of CD was 14.6% (141/969), while BPCD prevalence was 7.5%. Females had a higher prevalence than males: 17.8% vs. 9.8%, p < 0.001. The CD group had lower HbA1c and more frequent hypoglycemia than the seronegative group. Conclusions: This study highlighted the high prevalence of CD in T1D Saudi patients. CD has multiple effects on glycemic control, growth, and puberty in children and adolescents with T1D. We emphasize the importance of early screening for CD at the time of diabetes diagnosis and periodically after that or if any atypical features present, especially anemia, growth delay, underweight, or frequent hypoglycemia. Full article

Retinal vessel segmentation is crucial for diagnosing and monitoring ophthalmic and systemic diseases. Optical Coherence Tomography Angiography (OCTA) enables detailed imaging of the retinal microvasculature, but existing methods for OCTA segmentation face significant limitations, such as susceptibility to noise, difficulty in handling class imbalance, and challenges in accurately segmenting complex vascular morphologies. In this study, we propose VDMNet, a novel segmentation network designed to overcome these challenges by integrating several advanced components. Firstly, we introduce the Fast Multi-Head Self-Attention (FastMHSA) module to effectively capture both global and local features, enhancing the network’s robustness against complex backgrounds and pathological interference. Secondly, the Vessel Dynamic Convolution (VDConv) module is designed to dynamically adapt to curved and crossing vessels, thereby improving the segmentation of complex morphologies. Furthermore, we employ the Multi-Scale Fusion (MSF) mechanism to aggregate features across multiple scales, enhancing the detection of fine vessels while maintaining vascular continuity. Finally, we propose Weighted Asymmetric Focal Tversky Loss (WAFT Loss) to address class imbalance issues, focusing on the accurate segmentation of small and difficult-to-detect vessels. The proposed framework was evaluated on the publicly available ROSE-1 and OCTA-3M datasets. Experimental results demonstrated that our model effectively preserved the edge information of tiny vessels and achieved state-of-the-art performance in retinal vessel segmentation across several evaluation metrics. These improvements highlight VDMNet’s superior ability to capture both fine vascular details and overall vessel connectivity, making it a robust solution for retinal vessel segmentation. Full article

In high-altitude scenarios, targets tend to occupy a small number of pixels within the UAV’s field of view, resulting in substantial errors when identity recognition is attempted based solely on appearance features during multi-UAV joint tracking. Existing methodologies typically propose projecting multi-view data onto a single plane and leveraging distance information for identity association; however, their accuracy remains low as they are contingent on one-dimensional target information. To address this limitation, this paper introduces the UAVST-HM (UAV Swarm Tracking in High-altitude scenarios for Multiple targets) model, specifically designed to handle the characteristics of targets in the field of view of multiple UAVs at high altitudes. Initially, we develop techniques for extracting targets’ appearance, geometric, and distribution features. Subsequently, adaptive weights, calculated based on the mean of the respective features, are devised to amalgamate these diverse features, thereby constructing a cost matrix for cross-view target identity matching. This matrix is processed through the Hungarian algorithm, and multi-view target identity association is ultimately achieved via threshold filtering. On the MDMT dataset, our method enhances the MDA indicator, which assesses cross-view target identity matching, by 1.78 percentage points compared to the current state of the art. This significant enhancement substantially improves the overall efficacy of multi-UAV joint visual tracking from a high-altitude perspective. Full article

Brain–computer interfaces, where motor imagery electroencephalography (EEG) signals are transformed into control commands, offer a promising solution for enhancing the standard of living for disabled individuals. However, the performance of EEG classification has been limited in most studies due to a lack of attention to the complementary information inherent at different temporal scales. Additionally, significant inter-subject variability in sensitivity to biological motion poses another critical challenge in achieving accurate EEG classification in a subject-dependent manner. To address these challenges, we propose a novel machine learning framework combining multi-scale feature fusion, which captures global and local spatial information from different-sized EEG segmentations, and adaptive Lasso-based feature selection, a mechanism for adaptively retaining informative subject-dependent features and discarding irrelevant ones. Experimental results on multiple public benchmark datasets revealed substantial improvements in EEG classification, achieving rates of 81.36%, 75.90%, and 68.30% for the BCIC-IV-2a, SMR-BCI, and OpenBMI datasets, respectively. These results not only surpassed existing methodologies but also underscored the effectiveness of our approach in overcoming specific challenges in EEG classification. Ablation studies further confirmed the efficacy of both the multi-scale feature analysis and adaptive selection mechanisms. This framework marks a significant advancement in the decoding of motor imagery EEG signals, positioning it for practical applications in real-world BCIs. Full article

Trees and their morphology can mitigate the urban heat island (UHI) effect, but the impacts of tree species and their two-dimensional (2D) and three-dimensional (3D) morphological characteristics on the thermal environment of residential spaces at the building scale have not been effectively evaluated. This research extracted the data of trees in the spatial range of a 50 m radius of the sampling sites located in a subtropical humid city’s residential area based on unmanned aerial vehicle (UAV) imagery and field measurements. It included Ficus microcarpa L. f., Cinnamomum camphora (L.) J. Presl, and Alstonia scholaris (L.) R. Br. as three typical evergreen tree species and six quantitative indicators of trees, with the number of trees (N) serving as fundamental indicator and mean canopy width (MCW), mean canopy height (MCH), mean tree height (MTH), canopy biomass (CV), and mean canopy biomass (MCV) as morphological characteristic indicators. We analyzed the impact of the six indicators above on two thermal environment parameters: Air temperature (AT) and relative humidity (RH), by correlation analysis and multiple linear regression analysis. Results showed that: (1) F. microcarpa, as a dominant local species, provided more than 65% of the tree canopy volume within the study area (50 m radius buffer zones), and its contribution to cooling and humidification effects was superior to those of C. camphora and A. scholaris. (2) The MTH and CV of F. microcarpa are the key factors influencing daytime AT and RH, respectively, with temporal fluctuation in impact intensity during the spring (May) daytime. (3) The MTH and N of F. microcarpa show the best cooling effect (adjusted R² = 0.731, p < 0.05) during midday (13:00–14:00 p.m.), while its CV and MTH have the best humidification effect (adjusted R² = 0.748, p < 0.05) during the morning (9:00–10:00 a.m.) among three typical tree species. The 2D and 3D morphological characteristic indicators effectively describe the impact and variation of tree species on the spring microclimate within small-scale residential spaces. This work provides new insights into the thermal benefits brought by the spatial growth features of trees at the building scale and offers reference for urban residential areas in the planning and management related to tree species selection, canopy maintenance, and the improvement of thermal comfort for inhabitants. Full article

Human activity recognition (HAR) in real-world settings has gained significance due to the growth of Internet of Things (IoT) devices such as smartphones and smartwatches. Nonetheless, limitations such as fluctuating environmental conditions and intricate behavioral patterns have impacted the accuracy of the current procedures. This research introduces an innovative methodology employing a modified deep residual network, called 1D-ResNeXt, for IoT-enabled HAR in uncontrolled environments. We developed a comprehensive network that utilizes feature fusion and a multi-kernel block approach. The residual connections and the split–transform–merge technique mitigate the accuracy degradation and reduce the parameter number. We assessed our suggested model on three available datasets, mHealth, MotionSense, and Wild-SHARD, utilizing accuracy metrics, cross-entropy loss, and F1 score. The findings indicated substantial enhancements in proficiency in recognition, attaining 99.97% on mHealth, 98.77% on MotionSense, and 97.59% on Wild-SHARD, surpassing contemporary methodologies. Significantly, our model attained these outcomes with considerably fewer parameters (24,130–26,118) than other models, several of which exceeded 700,000 parameters. The 1D-ResNeXt model demonstrated outstanding effectiveness under various ambient circumstances, tackling a significant obstacle in practical HAR applications. The findings indicate that our modified deep residual network presents a viable approach for improving the dependability and usability of IoT-based HAR systems in dynamic, uncontrolled situations while preserving the computational effectiveness essential for IoT devices. The results significantly impact multiple sectors, including healthcare surveillance, intelligent residences, and customized assistive devices. Full article

The increasing use of electric vehicles has made fault diagnosis in electric drive motors, particularly in variable speed drives (VSDs) using three-phase induction motors, a critical area of research. This article presents a fault classification model based on machine learning (ML) algorithms to identify various faults under six operating conditions: normal operating mode (NOM), phase-to-phase fault (PTPF), phase-to-ground fault (PTGF), overloading fault (OLF), over-voltage fault (OVF), and under-voltage fault (UVF). A dataset simulating real-world operating conditions, consisting of 39,034 instances and nine key motor features, was analyzed. Comprehensive data preprocessing steps, including missing value removal, duplicate detection, and data transformation, were applied to enhance the dataset’s suitability for ML models. Yeo–Johnson and Hyperbolic Sine transformations were used to reduce skewness and improve the normality of the features. Multiple ML algorithms, including CatBoost, Random Forest (RF) Classifier, AdaBoost, and quadratic discriminant analysis (QDA), were trained and evaluated using Bayesian optimization with cross-validation. The CatBoost model achieved the best performance, with an accuracy of 94.1%, making it the most suitable model for fault classification in electric vehicle drive motors. Full article

The semantic segmentation of high-resolution remote sensing images (HRRSIs) faces persistent challenges in handling complex architectural structures and shadow occlusions, limiting the effectiveness of existing deep learning approaches. To address these limitations, we propose an attention-focused feature enhancement network (AFENet) with a novel encoder–decoder architecture. The encoder architecture combines ResNet50 with a parallel multistage feature enhancement group (PMFEG), enabling robust feature extraction through optimized channel reduction, scale expansion, and channel reassignment operations. Building upon this foundation, we develop a global multi-scale attention mechanism (GMAM) in the decoder that effectively synthesizes spatial information across multiple scales by learning comprehensive global–local relationships. The architecture is further enhanced by an efficient feature-weighted fusion module (FWFM) that systematically integrates remote spatial features with local semantic information to improve segmentation accuracy. Experimental results across diverse scenarios demonstrate that AFENet achieves superior performance in building structure detection, exhibiting enhanced segmentation connectivity and completeness compared to state-of-the-art methods. Full article

The maturation of internet usage environments has elevated User Experience (UX) to a critical factor in system success. However, traditional manual UX testing methods are hampered by subjectivity and lack of standardization, resulting in time-consuming and costly processes. This study explores the potential of Large Language Models (LLMs) to address these challenges by developing an automated UX testing tool. Our innovative approach integrates the Rapi web recording tool to capture user interaction data with the analytical capabilities of LLMs, utilizing Nielsen’s usability heuristics as evaluation criteria. This methodology aims to significantly reduce the initial costs associated with UX testing while maintaining assessment quality. To validate the tool’s efficacy, we conducted a case study featuring a tennis-themed course reservation system. The system incorporated multiple scenarios per page, allowing users to perform tasks based on predefined goals. We employed our automated UX testing tool to evaluate screenshots and interaction logs from user sessions. Concurrently, we invited participants to test the system and complete UX questionnaires based on their experiences. Comparative analysis revealed that varying prompts in the automated UX testing tool yielded different outcomes, particularly in detecting interface elements. Notably, our tool demonstrated superior capability in identifying issues aligned with Nielsen’s usability principles compared to participant evaluations. This research contributes to the field of UX evaluation by leveraging advanced language models and established usability heuristics. Our findings suggest that LLM-based automated UX testing tools can offer more consistent and comprehensive assessments. Full article

Background: Hepatocyte ballooning (HB) is a significant histological characteristic linked to the advancement of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH). Although clinicians now consider liver biopsy the most reliable method for identifying HB, its invasive nature and related dangers highlight the need for the development of non-invasive diagnostic options. Objective: This study aims to develop a novel methodology that combines deep learning and machine learning techniques to accurately identify and measure hepatobiliary abnormalities in liver ultrasound images. Methods: The research team expanded the dataset, consisting of ultrasound images, and used it for training deep convolutional neural networks (CNNs) such as InceptionV3, ResNet50, DenseNet121, and EfficientNetB0. A hybrid approach, combining InceptionV3 for feature extraction with a Random Forest classifier, emerged as the most accurate and stable method. An approach of dual dichotomy classification was used to categorize images into two stages: healthy vs. sick, and then mild versus severe ballooning.. Features obtained from CNNs were integrated with conventional machine learning classifiers like Random Forest and Support Vector Machines (SVM). Results: The hybrid approach achieved an accuracy of 97.40%, an area under the curve (AUC) of 0.99, and a sensitivity of 99% for the ‘Many’ class during the third phase of evaluation. The dual dichotomy classification enhanced the sensitivity in identifying severe instances of HB. The cross-validation process confirmed the strength and reliability of the suggested models. Conclusions: These results indicate that this combination method can decrease the need for invasive liver biopsies by providing a non-invasive and precise alternative for early identification and monitoring of NAFLD and NASH. Subsequent research will prioritize the validation of these models using larger datasets from multiple centers to evaluate their generalizability and incorporation into clinical practice. Full article

New programming languages are often designed to keep up with technological advancements and project requirements while also learning from previous attempts and introducing more powerful expression mechanisms. However, most existing dynamic programming languages rely on English keywords and lack features that facilitate easy translation of language syntax. Additionally, maintaining multiple implementations of the same language for different platforms, such as desktops and microcontrollers, can lead to inconsistencies and fragmented features. Furthermore, they usually do not use visual programming to fully implement the compiler and virtual machine. In this research paper, we introduce Ring—a dynamically-typed language with a lightweight implementation. However, it boasts several advantages, including a rich and versatile standard library and direct support for classes and object-oriented programming. The Ring language offers customization features. For instance, it allows easy modification of the language syntax multiple times, enabling programming by writing code using Arabic, English, or other keywords. Additionally, the language permits the creation of domain-specific languages through new features that extend object-oriented programming, allowing for specialized languages resembling CSS or Supernova. In the era of the Internet of Things, instead of creating another language implementation to support microcontrollers, the same Ring implementation allows us to create projects and applications for desktops, the web, WebAssembly, Android, or Raspberry Pi Pico. The Ring Compiler and Virtual Machine are designed using the PWCT Visual Programming language based on ANSI C. The visual implementation is composed of 18,945 components that generate 24,743 lines of code, which increases the abstraction level by approximately 23.5% and hides unnecessary details. Full article