检索结果-内蒙古大学图书馆

Research on recognition algorithm for special-shaped parts based on improved yolov5s

JOURNAL OF ELECTRONIC IMAGING 2024年第1期33卷 013050-013050页

作者： Zhang, Jiarui Guo, Zhongfeng Yang, Junlin Shenyang Univ Technol Sch Mech Engn Shenyang Peoples R China

With the rapid development of machinery and equipment modernization, more and more non-standard shaped parts are designed and put into specific occasions to use to meet the needs of special circumstances. Therefore, how to quickly recognize the shaped parts has become an urgent need for a technology. To recognize shaped parts, deep learning methods such as the widely used yolov5s network are commonly employed. However, directly deploying the official network model has drawbacks, including heavy reliance on data, poor detection results for small target objects, and high hardware requirements. These issues increase the threshold for non-professionals to use it. For this reason, this paper designs an improved network based on yolov5s. This paper proposes improvements in terms of both lightness and accuracy. In terms of light weight, the backbone of yolov5s is replaced by MobileNetV3;and the convolution and C3 module of the head part of yolov5s is replaced by phantom convolution and C3Ghost module, and the attention mechanism layer is trimmed to reduce the number of computational parameters and model size. In terms of accuracy, non-maximum suppression (NMs) is improved to soft-NMs;intersection over union (IoU) loss function is replaced with distance-IoU loss function. And trained on the homemade shaped parts dataset, the results show that the average accuracy of the improved network is 99.2% in the test case, the model size is 2.4M, and the detection time is 1.5 ms per image, which is a significant increase in speed and accuracy compared with other unmodified networks, and a substantial decrease in the model size and the number of parameters.

关键词： yolov5s algorithm machine vision special-shaped part recognition deep learning

来源：评论

学校读者我要写书评

暂无评论

Gesture recognition based on modified yolov5s

引用

IET IMAGE PROCEssING 2022年第8期16卷 2124-2132页

作者： Hu, Dunli Zhu, Jun Liu, Jiayu Wang, Jiaju Zhang, Xiaoping Beijing Key Lab Field Bus Technol & Automat Beijing 100144 Peoples R China North China Univ Technol Coll Elect & Control Engn Beijing 100144 Peoples R China

With the development of artificial intelligence technology, human-computer interaction technology through gestures, images and voices has gradually become a hot topic for discussion. A modified yolov5s gesture recognition method is proposed in the field of human-computer cooperation by optimizing the network structure of yolov5s backbone, CNN is replaced by Ghostbottleneck module to increase the target occlusion recognition rate. secondly, tensor stitching is added to the output of Ghostbottleneck module for up sampling to strengthen the reuse of image features. Finally, the detection ability of the improved model in the face of complex environment is verified on the self-made data set. Experimental results show that, the mAP@0.5 (mean average precision) of the modified yolov5s is 94.49%, the AP (average precision) is 94.2%. By comparing the yolov5s algorithm, yolov4 algorithm,yolov3 algorithm and ssD algorithm, the detection accuracy of the modified method has been significantly improved, which can fully meet the application requirements of real-time detection of gesture-controlled robots.

关键词： convolutional neural nets deep learning (artificial intelligence) gesture recognition target occlusion recognition rate object detection yolov5s algorithm yolov5s backbone human-computer cooperation gesture recognition method artificial intelligence technology modified yolov5 human-computer interaction technology gesture-controlled robots artificial intelligence Ghostbottleneck module feature extraction

来源：评论

学校读者我要写书评

暂无评论

Multi-target detection method for driving area of mine driverless rail locomotives based on lightweight network

引用

sIGNAL IMAGE AND VIDEO PROCEssING 2025年第4期19卷 1-10页

作者： Wang, Wenshan Hu, Kun Jiang, Hao Anhui Univ Sci & Technol Anhui Intelligent Mine Technol & Equipment Engn Re Huainan Peoples R China Anhui Univ Sci & Technol Sch Mechatron Engn Huainan Peoples R China Anhui Univ Sci & Technol Collaborat Innovat Ctr Min Intelligent Technol & E Huainan Peoples R China

Existing target detection algorithms have limitations in complex mine environments like low illumination, small targets, background interference, occlusion, and motion blur. Also, their complex network structures and large parameter volumes can't meet real-time detection needs of edge devices. Thus, a lightweight network-based multi-target detection method for mine driverless rail locomotive driving areas was proposed. A dataset of seven target images (electric locomotives, miners, etc.) in five scenarios (normal & low illumination, etc.) was constructed. Based on yolov5s, improvements were made: adding a small target detection layer to enhance small target detection;using the GhostBottleNeck module to replace BottleNeck in C3 to build C3Ghost, reducing calculation and parameters and compensating for the added layer;introducing the simAM attention mechanism to focus on targets and suppress interference;replacing CIoU with sIoU loss function to speed up convergence. Experimental results show the proposed lightweight network cuts parameters by 12.3%, boosts mAP by 1.7%, and is more suitable for multi-object detection in mine rail locomotive driving areas.

关键词： yolov5s algorithm small target detection layer C3Ghost simAM attention mechanism sIoU loss function

来源：评论

学校读者我要写书评

暂无评论

swinT-yolov5s:Improved yolov5s for Vehicle-mounted Infrared Target Detection 41

SwinT-YOLOv5s:Improved YOLOv5s for Vehicle-mounted Infrared ...

引用

第41届中国控制会议

作者： Xiuli Xin Feng Pan Jiacheng Wang Xiaoxue Feng Liwei shao School of Automation Beijing Institute of Technology Zhongshan Research Institute Beijing Institute of Technology

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

Infrared vehicle-mounted target detection is an important research direction in assisted driving,but also a very challenging *** infrared target detection methods often have problems such as high missed detection rate and false alarm in complex background,small target size and occlusion scene.A swinT-yolov5 s algorithm is proposed by the fusion of attention mechanism and convolutional *** on yolov5 s algorithm,a detection layer is added to enhance the detection ability of small target *** CBAM modules are inserted into the backbone network to make the model pay more attention to the useful information and resist the interference of redundant information,so as to improve the detection ability in dense *** addition,the swin Transfomer encoders are used to replace some part of C3 modules to improve the model's ability of mining potential feature details and further improve the detection accuracy of the *** results show that the improved algorithm increases the average precision(IOU=0.5) and precision rate by 5.60% and 4.20% compared with the original yolov5 s model,and has good generalization ability in remote small target and overlapping target scenarios.

关键词： Object Detection infrared image yolov5s algorithm CBAM swin Transformer Encoder

来源：评论

学校读者我要写书评

暂无评论

Real-time Vehicle Pedestrian Detection and Tracking algorithm based on Computer Vision 24

Real-time Vehicle Pedestrian Detection and Tracking Algorith...

引用

3rd International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML)

作者： Ye, Liping Lang Pei Wuhan Qingchuan Univ Coll Comp Sci Wuhan Hubei Peoples R China

ISBN: (纸本)9798400709777

Detection and tracking of vehicle pedestrians have important application values in the fields of intelligent driving and traffic monitoring. To this end, the study improves the yolov5s algorithm by replacing the backbone network of yolov5s with sGWin Transformer V2, and introduces the CBAM module, while optimizing the sIoU loss function to obtain the improved yolov5s algorithm. The improved yolov5s algorithm is used for the detection of vehicles and pedestrians in real-time video, then the fusion model is used to correlate the motion trajectories of the detected targets, and finally the Kalman filter tracking algorithm is applied to correct the tracking prediction results to realize the fast, accurate and continuous detection and tracking of vehicles and pedestrians. The results show that the tracking and detection accuracy of the method used in the study is 84.7%, and 51 vehicles and 32 pedestrians are accurately labeled. Research algorithms can accurately achieve vehicle and pedestrian recognition in complex road environments, and provide technical guidance for object recognition of the same type.

关键词： Vehicle pedestrians Real-time detection and tracking yolov5s algorithm Kalman filter tracking algorithm

来源：评论

学校读者我要写书评

暂无评论

Fish school Detection algorithm Improved Based on yolov5 3

Fish School Detection Algorithm Improved Based on Yolov5

引用

3rd International Conference on Electronic Information Engineering and Computer Communication, EIECC 2023

作者： Jiang, Zhiqiang Zhang, Weidong Jiang, Yong School of Computer Science and Technology Southwest University of Science and Technology Mianyang China

ISBN: (纸本)9798350359961

Water conservancy experts obtain fish school information through analyzing sonar videos, providing objective basis for the construction of fishway and guiding their construction. In order to reduce the missed detection rate of object detection in video analysis, a position attention module was introduced on yolov5s to enhance the expression ability of learning features in mobile networks;Introducing a coordinate attention module allows the network to understand the spatial relationships between objects in the image, perceive coordinate information to a certain extent, and improve detection performance. Final, the enhanced model outperforms the original yolov5s model by improving mAP0.5:0.95 by 1.3% and Recall by 2.58%. © 2023 IEEE.

关键词： coordinate convolution fish detection position attention yolov5s algorithm

来源：评论

学校读者我要写书评

暂无评论

study on Overheating Infrared Image Recognition of substation Insulated Casing based on Machine Learning

Study on Overheating Infrared Image Recognition of Substatio...

引用

2023 IEEE International Conference on Applied superconductivity and Electromagnetic Devices, AsEMD 2023

作者： Wang, sen Jiao, Tinghao Liang, Daili shi, Weiyang Zhang, Yang Xiao, Yanhui Shenyang Institute of Engineering Department of College of Automation Shenyang110000 China Shenyang Institute of Engineering Department of College of Electric Shenyang110000 China

ISBN: (纸本)9798350301571

As a more common fault type of electrical equipment, the economic loss of electrical equipment caused by the slow identification speed of overheating problems is severe. To solve the above issues, we choose yolov5s as the original network model, introduce deep separable convolution based on the original model, and add the simAM attention mechanism to enhance feature extraction. The results show that introducing deep separable convolution and simAM attention mechanism not only ensures the recognition accuracy but also improves the efficiency of training and recognition. The precision rate increases by 3.11% and the recall rate by 2.54%. This improvement provides a new idea for the identification and diagnosis of overheating of electrical equipment. © 2023 IEEE.

关键词： infrared imaging insulation casing overheating yolov5s algorithm

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：