相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Computer Science, Artificial Intelligence
Kun Wang et al.
Summary: This study presents a method based on YOLOv3, utilizing AS-CBAM and HDC to improve detection accuracy and achieve real-time warnings for completely occluded objects. It introduces DIOU-NMS and a tracking algorithm based on Kalman filtering and Hungarian matching to enhance the system's ability to recognize occluded objects. Compared to traditional YOLOv3, the proposed method increases mAP and shows a faster processing speed and significant improvement in accuracy on datasets focusing on occlusion conditions.
APPLIED INTELLIGENCE
(2022)
Article
Computer Science, Information Systems
Vinod Kumar et al.
Summary: The paper proposed a key distribution protocol based on the Chinese Remainder Theorem that significantly reduced computational complexity, increased security, improved efficiency in handling large groups, and effectively prevented illegal access.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Meng-ting Fang et al.
Summary: Examinations are a method to select talents, with a perfect invigilation strategy enhancing fairness. The improved YOLOv3 algorithm enables automatic detection of abnormal behavior in examination rooms, while the frame-alternate dual-thread method optimizes the detection process.
Article
Construction & Building Technology
Zhenwei Yu et al.
Summary: The YOLOv4-FPM model improves accuracy and speed by using focal loss and pruning algorithm to optimize the loss function and accelerate detection speed. It expands the detection range and is able to effectively detect cracks in images of different sizes.
AUTOMATION IN CONSTRUCTION
(2021)
Article
Chemistry, Analytical
Akmalbek Abdusalomov et al.
Summary: This study introduces a real-time high-speed fire detection method using deep learning, with a special convolutional neural network developed to detect fire regions using the YOLOv3 algorithm. By modifying the algorithm, rapid and high-precision detection of fires of various shapes and sizes, even at a distance of 50 m, was successfully achieved. Experimental results demonstrated seamless classification performance compared to conventional fire detection frameworks.
Article
Multidisciplinary Sciences
Yang Jie et al.
Summary: Ship detection and tracking in inland waterways is crucial for navigation safety. This paper introduces improvements to the YOLOv3 detection algorithm and its application in Deep SORT tracking algorithm, enhancing performance in ship detection and tracking. The enhanced algorithm shows increased precision and frame rate, with Deep SORT proving superior in complex scenarios.
Article
Multidisciplinary Sciences
Ce Guo et al.
Summary: In the electronics industry, the YOLOv4-tiny method is utilized to improve the detection of electronic components, and different network structures are built to enhance accuracy by adaptively integrating middle- and high-level features. The method is validated successfully on an electronic component dataset, showing improved accuracy and high detection speed compared to other mainstream algorithms.
SCIENTIFIC REPORTS
(2021)
Article
Engineering, Multidisciplinary
Myeongho Jeon et al.
Summary: Automation of fire detection systems can reduce loss of life and property, with recent studies showing that convolutional neural networks outperform conventional image processing methods. However, previous studies had limitations in classifying fire images of various sizes, which led to the proposal of a multi-scale prediction framework utilizing feature maps from all scales and a feature-squeeze block for effective information utilization. Extensive evaluations showed that this method outperformed state-of-the-art convolutional neural networks, with a 97.89% F1-score and 0.0227 false positive rate in multiple evaluations.
Article
Computer Science, Information Systems
Yunji Zhao et al.
Summary: The paper proposes a TADS algorithm for fire smoke detection, utilizing pre-trained networks to distinguish arbitrary objects of interest from unpredictable environments. The depthwise separable method improves the algorithm's speed for real-time smoke detection, outperforming current methods in terms of accuracy and speed.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Chemistry, Analytical
Hansen Liu et al.
Summary: To address the threat of drones intruding into high-security areas, a pruned-YOLOv4 model is proposed in this paper, which reduces model size and increases speed by pruning convolutional channels and shortcut layers, while improving the accuracy of small drone detection through special data augmentation methods.
Article
Multidisciplinary Sciences
Huixuan Fu et al.
Summary: This article proposes a YOLOv4 marine target detection method fused with a convolutional attention module, which has been experimentally shown to achieve better results in detection accuracy and real-time performance.
Article
Chemistry, Analytical
Addie Ira Borja Parico et al.
Summary: This study developed a real-time pear fruit counter using YOLOv4 and Deep SORT algorithm, finding a balance between accuracy, speed, and computational cost, as well as providing a method for choosing the most suitable model for agricultural science applications. YOLOv4-CSP was identified as the most accurate model, while YOLOv4-tiny was deemed more suitable for applications requiring lower speed and computational cost.
Article
Computer Science, Information Systems
Medhasvi Kulshreshtha et al.
Summary: The paper presents an innovative mechanical design for resilient Trash-Collecting Robots using the Rocker-bogie mechanism, alongside the exploration of various object detection models for trash detection. The Trash-Collecting Robot is fully autonomous, equipped with sensors and controlled by Raspberry Pi and Arduino for path-planning and motion. Among the models tested, YOLOv4-tiny was chosen for its optimal balance between mAP and detection time. The design was successfully simulated on different terrains, demonstrating expected behavior.
Article
Agriculture, Dairy & Animal Science
Anne K. Schuetz et al.
Summary: The use of computer vision technology for automatic monitoring of animal behavior can provide a more accurate and efficient analysis of animal activity and residency patterns, avoiding the time and cost consumption of traditional manual analysis.
Article
Remote Sensing
Subroto Singha et al.
Summary: Drones are becoming increasingly popular, leading to a need for automated drone detection systems to prevent misuse, with research showing superior performance in detection capabilities.
Article
Thermodynamics
Pu Li et al.
CASE STUDIES IN THERMAL ENGINEERING
(2020)
Article
Computer Science, Theory & Methods
Yuki Sakai et al.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2019)
Article
Construction & Building Technology
Y. W. Ng et al.
TUNNELLING AND UNDERGROUND SPACE TECHNOLOGY
(2019)
Article
Environmental Sciences
Jiangqiao Yan et al.
Article
Optics
Yi Zhang et al.
Article
Computer Science, Information Systems
Liang Qi et al.
Article
Engineering, Electrical & Electronic
Xiuling Zhang et al.
JOURNAL OF ELECTRONIC IMAGING
(2019)
Article
Oncology
Lei Wang et al.
WORLD JOURNAL OF SURGICAL ONCOLOGY
(2019)
Article
Chemistry, Analytical
Jun Sang et al.
Article
Geosciences, Multidisciplinary
Zamri Chik et al.
ARABIAN JOURNAL OF GEOSCIENCES
(2014)
