Search Results (3,795)

The fast advancement of unmanned aerial vehicle (UAV) technology has facilitated its use across a wide range of scenarios. Due to the high mobility and flexibility of drones, the images they capture often exhibit significant scale variations and severe object occlusions, leading to a high density of small objects. However, the existing object detection algorithms struggle with detecting small objects effectively in cross-scale detection scenarios. To overcome these difficulties, we introduce a new object detection model, RPS-YOLO, based on the YOLOv8 architecture. Unlike the existing methods that rely on traditional feature pyramids, our approach introduces a recursive feature pyramid (RFP) structure. This structure performs two rounds of feature extraction, and we reduce one downsampling step in the first round to enhance attention to small objects during cross-scale detection. Additionally, we design a novel attention mechanism that improves feature representation and mitigates feature degradation during convolution by capturing spatial- and channel-specific details. Another key innovation is the proposed Localization IOU (LIOU) loss function for bounding box regression, which accelerates the regression process by incorporating angular constraints. Experiments conducted on the VisDrone-DET2021 and UAVDT datasets show that RPS-YOLO surpasses YOLOv8s, with an mAP50 improvement of 8.2% and 3.4%, respectively. Our approach demonstrates that incorporating recursive feature extraction and exploiting detailed information for multi-scale detection significantly improves detection performance, particularly for small objects in UAV images. Full article

Due to a significant shortage of healthcare staff, medical facilities are increasingly challenged by the need to deploy current staff more intensively, which can lead to significant complications for patients and staff. Digital surgical assistance systems that track all instruments used in procedures can make a significant contribution to relieving the load on staff, increasing efficiency, avoiding errors and improving hygiene. Due to data safety concerns, laborious data annotation and the complexity of the scenes, as well as to increase prediction accuracy, the provision of synthetic data is key to enabling the wide use of artificial intelligence for object recognition and tracking in OR settings. In this study, a synthetic data generation pipeline is introduced for the detection of eight surgical instruments during open surgery. Using 3D models of the instruments, synthetic datasets consisting of color images and annotations were created. These datasets were used to train common object detection networks (YOLOv8) and compared against networks solely trained on real data. The comparison, conducted on two real image datasets with varying complexity, revealed that networks trained on synthetic data demonstrated better generalization capabilities. A sensitivity analysis showed that synthetic data-trained networks could detect surgical instruments even at higher occlusion levels than real data-trained networks. Additionally, 1920 datasets were generated using different parameter combinations to evaluate the impact of various settings on detection performance. Key findings include the importance of object visibility, occlusion, and the inclusion of occlusion objects in improving detection accuracy. The results highlight the potential of synthetic datasets to simulate real-world conditions, enhance network generalization, and address data shortages in specialized domains like surgical instrument detection. Full article

The recall rate of the original YOLOv4 model for detecting internal defects in aluminum alloy welds is relatively low. To address this issue, this paper introduces an enhanced model, YOLOv4-cs1. The improvements include optimizing the stacking method of residual blocks, modifying the activation functions for different convolutional layers, and eliminating the downsampling layer in the PANet (Pyramid Attention Network) to preserve edge information. Building on these enhancements, the YOLOv4-cs2 model further incorporates an improved Spatial Pyramid Pooling (SPP) module after the third and fourth residual blocks. The experimental results demonstrate that the recall rates for pore and slag inclusion detection using the YOLOv4-cs1 and YOLOv4-cs2 models increased by 28.9% and 16.6%, and 45% and 25.2%, respectively, compared to the original YOLOv4 model. Additionally, the mAP values for the two models are 85.79% and 87.5%, representing increases of 0.98% and 2.69%, respectively, over the original YOLOv4 model. Full article

Accurate, rapid, and intelligent stored-grain insect detection and counting are important for integrated pest management (IPM). Existing stored-grain insect pest detection models are often not suitable for detecting tiny insects on the surface of grain bulks and often require high computing resources and computational memory. Therefore, this study presents a YOLO-SGInsects model based on YOLOv8s for tiny stored-grain insect detection on the surface of grain bulk by adding a tiny object detection layer (TODL), adjusting the neck network with an asymptotic feature pyramid network (AFPN), and incorporating a hybrid attention transformer (HAT) module into the backbone network. The YOLO-SGInsects model was trained and tested using a GrainInsects dataset with images captured from granaries and laboratory. Experiments on the test set of the GrainInsects dataset showed that the YOLO-SGInsects achieved a stored-grain insect pest detection mean average precision (mAP) of 94.2%, with a counting root mean squared error (RMSE) of 0.7913, representing 2.0% and 0.3067 improvement over the YOLOv8s, respectively. Compared to other mainstream approaches, the YOLO-SGInsects model achieves better detection and counting performance and is capable of effectively handling tiny stored-grain insect pest detection in grain bulk surfaces. This study provides a technical basis for detecting and counting common stored-grain insect pests on the surface of grain bulk. Full article

In this paper, the Efficient Channel Attention (ECA) mechanism is incorporated at the terminal layer of the YOLOv10 backbone network to enhance the feature expression capability. In addition, Transformer is introduced into the C3 module in the feature extraction process to construct the C3TR module to replace the original C2F module as the deepening network extraction module. In this study, both the ECA mechanism and the self-attention mechanism of Transformer are thoroughly analyzed and integrated into YOLOv10. The C3TR module is used as an important part to deepen the effect of network extraction in backbone network feature extraction. The self-attention mechanism is used to model the long-distance dependency relationship, capture the global contextual information, make up for the limitation of the local sensory field, and enhance the feature expression capability. The ECA module is added to the end of the backbone to globally model the channels of the feature map, distribute channel weights more equitably, and enhance feature expression capability. Extensive experiments on the electrical equipment dataset have demonstrated the high accuracy of the method, with a mAP of 89.4% compared to the original model, representing an improvement of 3.2%. Additionally, the mAP@[0.5, 0.95] reaches 61.8%, which is 5.2% higher than that of the original model. Full article

Electric wheelchairs are the primary means of transportation that enable individuals with disabilities to move independently to their desired locations. This paper introduces a novel, low-cost smart wheelchair system designed to enhance the mobility of individuals with severe disabilities through hand gesture recognition. Additionally, the system aims to support low-income individuals who previously lacked access to smart wheelchairs. Unlike existing methods that rely on expensive hardware or complex systems, the proposed system utilizes an affordable webcam and an Nvidia Jetson Nano embedded computer to process and recognize six distinct hand gestures—“Forward 1”, “Forward 2”, “Backward”, “Left”, “Right”, and “Stop”—to assist with wheelchair navigation. The system employs the “You Only Look Once version 8n” (YOLOv8n) model, which is well suited for low-spec embedded computers, trained on a self-collected hand gesture dataset containing 12,000 images. The pre-processing phase utilizes the MediaPipe library to generate landmark hand images, remove the background, and then extract the region of interest (ROI) of the hand gestures, significantly improving gesture recognition accuracy compared to previous methods that relied solely on hand images. Experimental results demonstrate impressive performance, achieving 99.3% gesture recognition accuracy and 93.8% overall movement accuracy in diverse indoor and outdoor environments. Furthermore, this paper presents a control circuit system that can be easily installed on any existing electric wheelchair. This approach offers a cost-effective, real-time solution that enhances the autonomy of individuals with severe disabilities in daily activities, laying the foundation for the development of affordable smart wheelchairs. Full article

Pavement surface distresses are analyzed by transportation agencies to determine section performance across their pavement networks. To efficiently collect and evaluate thousands of lane-miles, automated processes utilizing image-capturing techniques and detection algorithms are applied to perform these tasks. However, the precision of this novel technology often leads to inaccuracies that must be verified by pavement engineers. Developments in artificial intelligence and machine learning (AI/ML) can aid in the progress of more robust and precise detection algorithms. Deep learning models are efficient for visual distress identification of pavement. With the use of 2D/3D pavement images, surface distress analysis can help train models to efficiently detect and classify surface distresses that may be caused by traffic loading, weather, aging, and other environmental factors. The formation of these distresses is developing at a higher rate in coastal regions, where extreme weather phenomena are more frequent and intensive. This study aims to develop a YOLOv5 model with 2D/3D images collected in the states of Louisiana, Mississippi, and Texas in the U.S. to establish a library of data on pavement sections near the Gulf of Mexico. Images with a resolution of 4096 × 2048 are annotated by utilizing bounding boxes based on a class list of nine distress and non-distress objects. Along with emphasis on efforts to detect cracks in the presence of background noise on asphalt pavements, six scenarios for augmentation were made to evaluate the model’s performance based on flip probability in the horizontal and vertical directions. The YOLOv5 models are able to detect defined distresses consistently, with the highest mAP50 scores ranging from 0.437 to 0.462 throughout the training scenarios. Full article

Classroom behavior detection is a key task in constructing intelligent educational environments. However, the existing models are still deficient in detail feature capture capability, multi-layer feature correlation, and multi-scale target adaptability, making it challenging to realize high-precision real-time detection in complex scenes. This paper proposes an improved classroom behavior detection algorithm, YOLO-AMM, to solve these problems. Firstly, we constructed the Adaptive Efficient Feature Fusion (AEFF) module to enhance the fusion of semantic information between different features and improve the model’s ability to capture detailed features. Then, we designed a Multi-dimensional Feature Flow Network (MFFN), which fuses multi-dimensional features and enhances the correlation information between features through the multi-scale feature aggregation module and contextual information diffusion mechanism. Finally, we proposed a Multi-Scale Perception and Fusion Detection Head (MSPF-Head), which significantly improves the adaptability of the head to different scale targets by introducing multi-scale feature perception, feature interaction, and fusion mechanisms. The experimental results showed that compared with the YOLOv8n model, YOLO-AMM improved the mAP0.5 and mAP0.5-0.95 by 3.1% and 4.0%, significantly improving the detection accuracy. Meanwhile, YOLO-AMM increased the detection speed (FPS) by 12.9 frames per second to 169.1 frames per second, which meets the requirement for real-time detection of classroom behavior. Full article

This study addresses the challenges of detecting inferior fruits in table grapes in natural environments, focusing on subtle appearance differences, occlusions, and fruit overlaps. We propose an enhanced green grape fruit disease detection model named RF-YOLOv7. The model is trained on a dataset comprising images of small fruits, sunburn, excess grapes, fruit fractures, and poor-quality grape bunches. RF-YOLOv7 builds upon the YOLOv7 architecture by integrating four Contextual Transformer (CoT) modules to improve target-detection accuracy, employing the Wise-IoU (WIoU) loss function to enhance generalization and overall performance, and introducing the Bi-Former attention mechanism for dynamic query awareness sparsity. The experimental results demonstrate that RF-YOLOv7 achieves a detection accuracy of 83.5%, recall rate of 76.4%, mean average precision (mAP) of 80.1%, and detection speed of 58.8 ms. Compared to the original YOLOv7, RF-YOLOv7 exhibits a 3.5% increase in mAP, with only an 8.3 ms increase in detection time. This study lays a solid foundation for the development of automatic detection equipment for intelligent grape pruning. Full article

Industrial heat sources (IHSs) are major contributors to energy consumption and environmental pollution, making their accurate detection crucial for supporting industrial restructuring and emission reduction strategies. However, existing models either focus on single-class detection under complex backgrounds or handle multiclass tasks for simple targets, leaving a gap in effective multiclass detection for complex scenarios. To address this, we propose a novel multiclass IHS detection model based on the YOLOv8-FC framework, underpinned by the multiclass IHS training dataset constructed from optical remote sensing images and point-of-interest (POI) data firstly. This dataset incorporates five categories: cement plants, coke plants, coal mining areas, oil and gas refineries, and steel plants. The proposed YOLOv8-FC model integrates the FasterNet backbone and a Coordinate Attention (CA) module, significantly enhancing feature extraction, detection precision, and operational speed. Experimental results demonstrate the model’s robust performance, achieving a precision rate of 92.3% and a recall rate of 95.6% in detecting IHS objects across diverse backgrounds. When applied in the Beijing–Tianjin–Hebei (BTH) region, YOLOv8-FC successfully identified 429 IHS objects, with detailed category-specific results providing valuable insights into industrial distribution. It shows that our proposed multiclass IHS detection model with the novel YOLOv8-FC approach could effectively and simultaneously detect IHS categories under complex backgrounds. The IHS datasets derived from the BTH region can support regional industrial restructuring and optimization schemes. Full article

Mountainous roads are severely affected by environmental factors such as insufficient lighting and shadows from tree branches, which complicates the detection of drivers’ facial features and the determination of fatigue states. An improved method for recognizing driver fatigue states on mountainous roads using the YOLOv5 neural network is proposed. Initially, modules from Deformable Convolutional Networks (DCNs) are integrated into the feature extraction stage of the YOLOv5 framework to improve the model’s flexibility in recognizing facial characteristics and handling postural changes. Subsequently, a Triplet Attention (TA) mechanism is embedded within the YOLOv5 network to bolster image noise suppression and improve the network’s robustness in recognition. Finally, the Wing loss function is introduced into the YOLOv5 model to heighten the sensitivity to micro-features and enhance the network’s capability to capture details. Experimental results demonstrate that the modified YOLOv5 neural network achieves an average accuracy rate of 85% in recognizing driver fatigue states. Full article

In response to the need for deploying the YOLOv4-Tiny model on resource-constrained Field-Programmable Gate Array (FPGA) platforms for rapid inference, this study proposes a general optimization acceleration strategy and method aimed at achieving fast inference for object detection networks. This approach centers on the synergistic effect of several key strategies: a refined resource management strategy that dynamically adjusts FPGA hardware resource allocation based on the network architecture; a dynamic dual-buffering strategy that maximizes the parallelism of data computation and transmission; an interface access latency pre-configuration strategy that effectively improves data throughput; and quantization operations for dynamic bit width tuning of model parameters and cached variables. Experimental results on the ZYNQ7020 platform demonstrate that this accelerator operates at a frequency of 200 MHz, achieving an average computing performance of 36.97 Giga Operations Per Second (GOPS) with an energy efficiency of 8.82 Giga Operations Per Second per Watt (GOPS/W). Testing with a metal surface defect dataset maintains an accuracy of approximately 90% per image, while reducing the inference delay per frame to 185 ms, representing a 52.2% improvement in inference speed. Compared to other FPGA accelerator designs, the accelerator design strategies and methods proposed in this study showcase significant enhancements in average computing performance, energy efficiency, and inference latency. Full article

With the rapid development of machine learning and deep learning technology, cow face detection technology has achieved remarkable results. Traditional contact cattle identification methods are costly; are easy to lose and tamper with; and can lead to a series of security problems, such as untimely disease prevention and control, incorrect traceability of cattle products, and fraudulent insurance claims. In order to solve these problems, this study explores the application of cattle face detection technology in cattle individual detection to improve the accuracy of detection, an approach that is particularly important in smart animal husbandry and animal behavior analysis. In this paper, we propose a novel cow face detection network based on YOLOv7 improvement, named CFR-YOLO. First of all, the method of extracting the features of a cow’s face (including nose, eye corner, and mouth corner) is constructed. Then, we calculate the frame center of gravity and frame size based on these feature points to design the cow face detection CFR-YOLO network model. To optimize the performance of the model, the activation function of FReLU is used instead of the original SiLU activation function, and the CBS module is replaced by the CBF module. The RFB module is introduced in the backbone network; and in the head layer, the CBAM convolutional attention module is introduced. The performance of CFR-YOLO is compared with other mainstream deep learning models (including YOLOv7, YOLOv5, YOLOv4, and SSD) on a self-built cow face dataset. Experiments indicate that the CFR-YOLO model achieves 98.46% accuracy (precision), 97.21% recall (recall), and 96.27% average accuracy (mAP), proving its excellent performance in the field of cow face detection. In addition, comparative analyses with the other four methods show that CFR-YOLO exhibits faster convergence speed while ensuring the same detection accuracy; and its detection accuracy is higher under the condition of the same model convergence speed. These results will be helpful to further develop the cattle identification technique. Full article

The ever-growing volume of global electronic waste (e-waste) poses significant environmental and health challenges. Printed circuit boards (PCBs), which form the core of most electronic devices, contain valuable metals as well as hazardous materials. The efficient disassembly and recycling of e-waste is critical for both economic and environmental sustainability. The traditional manual disassembly methods are time-consuming, labor-intensive, and often hazardous. The integration of edge computing and the Internet of Things (IoT) provides a novel approach to automating the disassembly process, potentially transforming the way e-waste is managed. Automated disassembly of WPCBs involves the use of advanced technologies, specifically edge computing and the IoT, to streamline the recycling process. This strategy aims to improve the efficiency and sustainability of e-waste management by leveraging real-time data analytics and intelligent decision-making at the edge of the network. This paper explores the application of edge computing and the IoT in the automated disassembly of WPCBs, discussing the technological framework, benefits, challenges, and future prospects. The experimental results show that the YOLOv10 model achieves 99.9% average precision (AP), enabling accurate real-time detection of electronic components, which greatly facilitates the automated disassembly process. Full article

Currently, the identification of unsafe behaviors among construction workers predominantly relies on manual methods, which are time-consuming, labor intensive, and inefficient. To enhance identification accuracy and ensure real-time performance, this paper proposes an enhanced YOLOv5s framework with three strategic improvements: (1) adoption of the Focal-EIoU loss function to resolve sample imbalance and localization inaccuracies in complex scenarios; (2) integration of the Coordinate Attention (CA) mechanism, which enhances spatial perception through channel-direction feature encoding, outperforming conventional SE blocks in positional sensitivity; and (3) development of a dedicated small-target detection layer to capture critical fine-grained features. Based on the improved model, a method for identifying unsafe behaviors of construction workers is proposed. Validated through a sluice renovation project in Jiangsu Province, the optimized model demonstrates a 3.6% higher recall (reducing missed detections) and a 2.2% mAP improvement over baseline, while maintaining a 42 FPS processing speed. The model effectively identifies unsafe behaviors at water conservancy construction sites, accurately detecting relevant unsafe actions, while meeting real-time performance requirements. Full article