Search Results (5,353)

The speedy progress of computer vision and machine learning engineering has inaugurated novel means for improving the purchasing experiment in brick-and-mortar stores. This paper examines the utilization of YOLOv (You Only Look Once) and DeepSORT (Deep Simple Online and Real-Time Tracking) algorithms for the real-time detection and analysis of the purchasing penchant in brick-and-mortar market surroundings. By leveraging these algorithms, stores can track customer behavior, identify popular products, and monitor high-traffic areas, enabling businesses to adapt quickly to customer preferences and optimize store layout and inventory management. The methodology involves the integration of YOLOv5 for accurate and rapid object detection combined with DeepSORT for the effective tracking of customer movements and interactions with products. Information collected in in-store cameras and sensors is handled to detect tendencies in customer behavior, like repeatedly inspected products, periods expended in specific intervals, and product handling. The results indicate a modest improvement in customer engagement, with conversion rates increasing by approximately 3 percentage points, and a decline in inventory waste levels, from 88% to 75%, after system implementation. This study provides essential insights into the further integration of algorithm technology in physical retail locations and demonstrates the revolutionary potential of real-time behavior tracking in the retail industry. This research determines the foundation for future developments in functional strategies and customer experience optimization by offering a solid framework for creating intelligent retail systems. Full article

The detection of the human mental fatigue state holds immense significance due to its direct impact on work efficiency, specifically in system operation control. Numerous approaches have been proposed to address the challenge of fatigue detection, aiming to identify signs of fatigue and alert the individual. This paper introduces an approach to human mental fatigue assessment based on the application of machine learning techniques to the video of a working operator. For validation purposes, the approach was applied to a dataset, “Human Fatigue Assessment Based on Video Data” (HFAVD) integrating video data with features computed by using our computer vision deep learning models. The incorporated features encompass head movements represented by Euler angles (roll, pitch, and yaw), vital signs (blood pressure, heart rate, oxygen saturation, and respiratory rate), and eye and mouth states (blinking and yawning). The integration of these features eliminates the need for the manual calculation or detection of these parameters, and it obviates the requirement for sensors and external devices, which are commonly employed in existing datasets. The main objective of our work is to advance research in fatigue detection, particularly in work and academic settings. For this reason, we conducted a series of experiments by utilizing machine learning techniques to analyze the dataset and assess the fatigue state based on the features predicted by our models. The results reveal that the random forest technique consistently achieved the highest accuracy and F1-score across all experiments, predominantly exceeding 90%. These findings suggest that random forest is a highly promising technique for this task and prove the strong connection and association among the predicted features used to annotate the videos and the state of fatigue. Full article

Underwater vision data facilitate a variety of underwater operations, including underwater ecosystem monitoring, topographical mapping, mariculture, and marine resource exploration. Conventional laser-based underwater imaging systems with complex system architecture rely on high-cost laser systems with high power, and software-based methods can not enrich the physical information captured by cameras. In this manuscript, a low-cost modulated laser-based imaging system is proposed with a spot in the shape of a square ring to eliminate the overlap between the illumination light path and the imaging path, which could reduce the negative effect of backscatter on the imaging process and enhance imaging quality. The imaging system is able to achieve underwater imaging at long distance (e.g., 10 m) with turbidity in the range of 2.49 to 7.82 NTUs, and the adjustable divergence angle of the laser tubes enables the flexibility of the proposed system to image on the basis of application requirements, such as the overall view or partial detail information of targets. Compared with a conventional underwater imaging camera (DS-2XC6244F, Hikvision, Hangzhou, China), the developed system could provide better imaging performance regarding visual effects and quantitative evaluation (e.g., UCIQUE and IE). Through integration with the CycleGAN-based method, the imaging results can be further improved, with the UCIQUE increased by 0.4. The proposed low-cost imaging system with a compact system structure and low consumption of energy could be equipped with platforms, such as underwater robots and AUVs, to facilitate real-world underwater applications. Full article

Mildew in maize kernels is typically caused by various fungi, necessitating prompt detection and treatment to minimize losses during harvest and storage. In this study, a deep learning YOLOv5s algorithm based on machine vision technology was employed to develop a maize seed surface mildew detection model and to enhance its portability for deployment on additional mobile devices. To guarantee the fruitful progression of this research, an initial experiment was conducted on maize seeds to obtain a sufficient number of images of mildewed maize kernels, which were classified into three grades (sound, mild, and severe). Subsequently, a maize seed image was extracted to create an image of a single maize seed, which was then divided to establish the data set. An enhanced YOLOv5s–ShuffleNet–CBAM model was ultimately developed. The results demonstrated that the model achieved with an mAP50 value of 0.955 and a model size of 2.4 MB. This resulted in a notable reduction in the model parameters and calculation amount while simultaneously enhancing model precision. Furthermore, K-fold cross-validation demonstrated the model stability, and Grad-CAM validated the model effectiveness. In the future, the proposed lightweight model in this study can be applied to other crops in the context of portable or online inspection systems, thus advancing effective and high-quality agricultural applications. Full article

This study introduces Hybrid-DC, a hybrid deep-learning model integrating ResNet-50 and Vision Transformer (ViT) for high-accuracy steel surface defect classification. Hybrid-DC leverages ResNet-50 for efficient feature extraction at both low and high levels and utilizes ViT’s global context learning to enhance classification precision. A unique hybrid attention layer and an attention fusion mechanism enable Hybrid-DC to adapt to the complex, variable patterns typical of steel surface defects. Experimental evaluations demonstrate that Hybrid-DC achieves substantial accuracy improvements and significantly reduced loss compared to traditional models like MobileNetV2 and ResNet, with a validation accuracy reaching 0.9944. The results suggest that this model, characterized by rapid convergence and stable learning, can be applied for real-time quality control in steel manufacturing and other high-precision industries, enhancing automated defect detection efficiency. Full article

Improving the three-dimensional reconstruction of underwater scenes is a challenging and hot topic in the field of underwater robot vision system research. High dynamic interference underwater has always been one of the key issues affecting the 3D reconstruction of underwater scenes. However, due to the complex underwater environment and insufficient light, existing target detection algorithms cannot meet the requirements. This paper uses the YOLOv8 network as the basis of the algorithm and proposes an underwater dynamic target detection algorithm based on improved YOLOv8. This algorithm first improves the feature extraction layer of the YOLOv8 network, improves the convolutional network structure of Bottleneck, reduces the amount of calculation and improves detection accuracy. Secondly, it adds an improved SE attention mechanism to make the network have a better feature extraction effect; in addition, the confidence box loss function of the network is improved, and the CIoU loss function is replaced by the MPDIoU loss function, which effectively improves the model convergence speed. Experimental results show that the mAP value of the improved YOLOv8 underwater dynamic target detection algorithm proposed in this article can reach 95.1%, and it can detect underwater dynamic targets more accurately, especially small dynamic targets in complex underwater scenes. Full article

Deep learning has made significant strides, driving advances in areas like computer vision, natural language processing, and autonomous systems. In this paper, we further investigate the implications of the role of additive shortcut connections, focusing on models such as ResNet, Vision Transformers (ViTs), and MLP-Mixers, given that they are essential in enabling efficient information flow and mitigating optimization challenges such as vanishing gradients. In particular, capitalizing on our recent information bottleneck approach, we analyze how additive shortcuts influence the fitting and compression phases of training, crucial for generalization. We leverage Z-X and Z-Y measures as practical alternatives to mutual information for observing these dynamics in high-dimensional spaces. Our empirical results demonstrate that models with identity shortcuts (ISs) often skip the initial fitting phase and move directly into the compression phase, while non-identity shortcut (NIS) models follow the conventional two-phase process. Furthermore, we explore how IS models are still able to compress effectively, maintaining their generalization capacity despite bypassing the early fitting stages. These findings offer new insights into the dynamics of shortcut connections in neural networks, contributing to the optimization of modern deep learning architectures. Full article

Invasive aspergillosis is a rare but severe fungal infection primarily affecting immunocompromised individuals. The Coronavirus Disease-2019 (COVID-19) pandemic has introduced new complexities in managing aspergillosis due to the widespread use of corticosteroids for treating COVID-19-related respiratory distress, which can increase susceptibility to fungal infections. Here, we present a challenging case of progressive cerebral aspergillosis complicated by cavernous sinus thrombosis (CST) in a 67-year-old male with a history of COVID-19. The patient, initially misdiagnosed with temporal arteritis, received pulse corticosteroid therapy twice before presenting with persistent left-sided headaches and vision loss. Cranial imaging revealed findings consistent with fungal sinusitis, Tolosa–Hunt syndrome, and orbital pseudotumor, which progressed despite initial antifungal therapy. Subsequent magnetic resonance imaging indicated an invasive mass extending into the left cavernous sinus and other intracranial structures, raising suspicion of aspergillosis. A transsphenoidal biopsy confirmed Aspergillus infection, leading to voriconazole therapy. Despite aggressive treatment, follow-up imaging revealed significant progression, with extension to the right frontal region and left cavernous sinus. The patient then developed visual impairment in the right eye and was diagnosed with CST secondary to fungal sinusitis. Management included a combination of systemic antifungals and antibiotics; however, the patient declined surgical intervention. This case underscores the diagnostic challenges and rapid progression associated with cerebral aspergillosis in post-COVID-19 patients treated with corticosteroids. This report highlights the need for heightened clinical suspicion and prompt, targeted interventions in similar cases to improve patient outcomes. Further research is required to understand the optimal management of invasive fungal infections. Full article

Understanding activity levels during fasting is important for promoting healthy fasting practices. While most existing studies focus on step counts to objectively assess the impact of fasting on activity levels and behavioral changes, the results have been mixed. Despite evidence showing that individuals spend a significant amount of time sitting while fasting, there has been no objective measurement of body movement or activity levels during sitting and fasting. This research employs a video-based, unobtrusive human body movement measurement system to monitor upper body movements during fasting and non-fasting periods over several days. Key movement features, such as inactivity, movement speed, and movement scale, were automatically extracted from the video monitoring data using a computer vision pipeline. These features were then statistically compared using t tests between fasting and non-fasting periods, analyzed by hour of the day and across different days. The results of the monitoring of five participants during typical daily sitting office work and fasting for 3–5 days indicate no consistent pattern of upper body movement changes due to fasting among the participants. Full article

The advancement of unloading technology in combine harvesting is crucial for the intelligent development of agricultural machinery. Accurately measuring material pile height in transport vehicles is essential, as uneven accumulation can lead to spillage and voids, reducing loading efficiency. Relying solely on manual observation for measuring stack height can decrease harvesting efficiency and pose safety risks due to driver distraction. This research applies binocular vision to agricultural harvesting, proposing a novel method that uses a stereo matching algorithm to measure material pile height during harvesting. By comparing distance measurements taken in both empty and loaded states, the method determines stack height. A linear regression model processes the stack height data, enhancing measurement accuracy. A binocular vision system was established, applying Zhang’s calibration method on the MATLAB (R2019a) platform to correct camera parameters, achieving a calibration error of 0.15 pixels. The study implemented block matching (BM) and semi-global block matching (SGBM) algorithms using the OpenCV (4.8.1) library on the PyCharm (2020.3.5) platform for stereo matching, generating disparity, and pseudo-color maps. Three-dimensional coordinates of key points on the piled material were calculated to measure distances from the vehicle container bottom and material surface to the binocular camera, allowing for the calculation of material pile height. Furthermore, a linear regression model was applied to correct the data, enhancing the accuracy of the measured pile height. The results indicate that by employing binocular stereo vision and stereo matching algorithms, followed by linear regression, this method can accurately calculate material pile height. The average relative error for the BM algorithm was 3.70%, and for the SGBM algorithm, it was 3.35%, both within the acceptable precision range. While the SGBM algorithm was, on average, 46 ms slower than the BM algorithm, both maintained errors under 7% and computation times under 100 ms, meeting the real-time measurement requirements for combine harvesting. In practical operations, this method can effectively measure material pile height in transport vehicles. The choice of matching algorithm should consider container size, material properties, and the balance between measurement time, accuracy, and disparity map completeness. This approach aids in manual adjustment of machinery posture and provides data support for future autonomous master-slave collaborative operations in combine harvesting. Full article

In recent years, the field of deep learning and computer vision has increasingly focused on the problem of vehicle target detection, becoming the forefront of many technological innovations. YOLOv8, as an efficient vehicle target detection model, has achieved good results in many scenarios. However, when faced with complex traffic scenarios, such as occluded targets, small target detection, changes in lighting, and variable weather conditions, YOLOv8 still has insufficient detection accuracy and robustness. To address these issues, this paper delves into the optimization strategies of YOLOv8 in the field of vehicle target detection, focusing on the EMA module in the backbone part and replacing the original SPPF module with focal modulation technology, all of which effectively improved the model’s performance. At the same time, modifications to the head part were approached with caution to avoid unnecessary interference with the original design. The experiment used the UA-DETRAC dataset, which contains a variety of traffic scenarios, a rich variety of vehicle types, and complex dynamic environments, making it suitable for evaluating and validating the performance of traffic monitoring systems. The 5-fold cross-validation method was used to ensure the reliability and comprehensiveness of the evaluation results. The final results showed that the improved model’s precision rate increased from 0.859 to 0.961, the recall rate from 0.83 to 0.908, and the mAP50 from 0.881 to 0.962. Meanwhile, the optimized YOLOv8 model demonstrated strong robustness in terms of detection accuracy and the ability to adapt to complex environments. Full article

Drowning is among the most common reasons for children’s death aged one to fourteen around the globe, ranking as the third leading cause of unintentional injury death. With rising populations and the growing popularity of swimming pools in hotels and villas, the incidence of drowning has accelerated. Accordingly, the development of systems for detecting and preventing drowning has become increasingly critical to provide safe swimming settings. In this paper, we propose a comprehensive review of recent existing advancements in automated drowning detection and prevention systems. The existing approaches can be broadly categorized according to their objectives into two main groups: detection-based systems, which alert lifeguards or parents to perform manual rescues, and detection and rescue-based systems, which integrate detection with automatic rescue mechanisms. Automatic drowning detection approaches could be further categorized into computer vision-based approaches, where camera-captured images are analyzed by machine learning algorithms to detect instances of drowning, and sensing-based approaches, where sensing instruments are attached to swimmers to monitor their physical parameters. We explore the advantages and limitations of each approach. Additionally, we highlight technical challenges and unresolved issues related to this domain, such as data imbalance, accuracy, privacy concerns, and integration with rescue systems. We also identify future research opportunities, emphasizing the need for more advanced AI models, uniform datasets, and better integration of detection with autonomous rescue mechanisms. This study aims to provide a critical resource for researchers and practitioners, facilitating the development of more effective systems to enhance water safety and minimize drowning incidents. Full article

Thailand’s transition to a low-carbon economy faces significant challenges, including a dependency on fossil fuels, fluctuating energy costs, and limited policy clarity. This study conducts a comparative analysis of energy transition policies in Germany, Japan, Australia, Malaysia, and Singapore to derive actionable lessons that can be adapted to Thailand’s socio-economic and energy contexts. Using the Integrated National Energy Planning (INEP) framework and Network Governance Theory, the research identifies key strategies, such as setting clear and achievable renewable energy targets, establishing robust legal frameworks, fostering multi-stakeholder engagement, and encouraging decentralized governance. The findings highlight the importance of long-term vision, inclusive governance, and targeted investments in renewable technologies to accelerate energy transitions. This paper presents policy guidelines to enhance Thailand’s energy security and contribute to its climate goals by promoting public awareness and strengthening institutional capacities. By adapting these strategies, Thailand can align with global energy trends, reduce its reliance on fossil fuels, and advance toward a resilient and sustainable energy system, aligned with global energy trends while addressing its unique socio-economic context. Full article

Deep learning has dramatically advanced computer vision tasks, including person re-identification (re-ID), substantially improving matching individuals across diverse camera views. However, person re-ID systems remain vulnerable to adversarial attacks that introduce imperceptible perturbations, leading to misidentification and undermining system reliability. This paper addresses the challenge of robust person re-ID in the presence of adversarial examples by estimating attack intensity to enable effective detection and adaptive purification. The proposed approach leverages the observation that adversarial examples in retrieval tasks disrupt the relevance and internal consistency of retrieval results, degrading re-ID accuracy. This approach estimates the attack intensity and dynamically adjusts the purification strength by analyzing the query response data, addressing the limitations of fixed purification methods. This approach also preserves the performance of the model on clean data by avoiding unnecessary manipulation while improving the robustness of the system and its reliability in the presence of adversarial examples. The experimental results demonstrate that the proposed method effectively detects adversarial examples and estimates the attack intensity through query response analysis. This approach enhances purification performance when integrated with adversarial purification techniques in person re-ID systems. Full article

To improve the automation level of highway maintenance operations, the lightweight YOLOv5-Lite-s neural network was deployed in embedded devices to assist an automatic traffic cone retractor in completing recognition and positioning operations. The system used the lightweight shuffle Net network as a backbone for feature extraction, replaced convolutional layers with focus modules to reduce computational complexity, and reduced the use of the C3 layer to increase network speed, thereby meeting the speed and accuracy requirements of traffic cone placement and retraction operations while maintaining acceptable model inference accuracy. The experimental results show that the network could maintain recognition accuracy and speed values of around 89% and 9 fps under different working conditions such as varying distances, lighting conditions, and occlusions, meeting the technical requirements for deploying and retrieving cones at a speed of 30 cones per minute when the operating vehicle’s speed was 20 km/h. The automatic traffic cone placement and retraction system operated accurately and stably, achieving the application of machine vision in traffic cone retraction operations. Full article