Camouflaged Object Detection Based on Improved YOLO v5 Algorithm
Since the camouflage object is highly similar to the surrounding environment with a rather small size,the general detection algorithm is not fully applicable to the camouflaged object detection task,which makes the detection of camouflaged object more challenging than the general detection task.In o...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial office of Computer Science
2021-10-01
|
| Series: | Jisuanji kexue |
| Subjects: | |
| Online Access: | http://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2021-10-226.pdf |
| _version_ | 1819021651030835200 |
|---|---|
| author | WANG Yang, CAO Tie-yong, YANG Ji-bin, ZHENG Yun-fei, FANG Zheng, DENG Xiao-tong, WU Jing-wei, LIN Jia |
| author_facet | WANG Yang, CAO Tie-yong, YANG Ji-bin, ZHENG Yun-fei, FANG Zheng, DENG Xiao-tong, WU Jing-wei, LIN Jia |
| author_sort | WANG Yang, CAO Tie-yong, YANG Ji-bin, ZHENG Yun-fei, FANG Zheng, DENG Xiao-tong, WU Jing-wei, LIN Jia |
| collection | DOAJ |
| description | Since the camouflage object is highly similar to the surrounding environment with a rather small size,the general detection algorithm is not fully applicable to the camouflaged object detection task,which makes the detection of camouflaged object more challenging than the general detection task.In order to solve this problem,the existing methods are analyzed in this paper and a detection algorithm for camouflage object is proposed based on the YOLO v5 algorithm.A new feature extraction network combined with attention mechanism is designed to highlight the feature information of the camouflage target.The original path aggregation network is improved so that the high,middle and lowly level feature map information is fully fused.The semantic information of the target is strengthened by nonlinear pool module,and the detection feature map size is increased to improve the detection recall rate of the small size target.On a public camouflage target dataset,the proposed algorithm is tested with 7 algorithms.The mAP of the proposed algorithm is 4.4% higher than that of the original algorithm,while the recall rate has improved 2.8%,which verifies the effectiveness of the algorithm for camouflaged object detection and the great advantage in accuracy compared with other algorithms.<br/> |
| first_indexed | 2024-12-21T04:10:29Z |
| format | Article |
| id | doaj.art-36d82f61cbba4443acea541abcb95757 |
| institution | Directory Open Access Journal |
| issn | 1002-137X |
| language | zho |
| last_indexed | 2024-12-21T04:10:29Z |
| publishDate | 2021-10-01 |
| publisher | Editorial office of Computer Science |
| record_format | Article |
| series | Jisuanji kexue |
| spelling | doaj.art-36d82f61cbba4443acea541abcb957572022-12-21T19:16:28ZzhoEditorial office of Computer ScienceJisuanji kexue1002-137X2021-10-01481022623210.11896/jsjkx.210100058Camouflaged Object Detection Based on Improved YOLO v5 AlgorithmWANG Yang, CAO Tie-yong, YANG Ji-bin, ZHENG Yun-fei, FANG Zheng, DENG Xiao-tong, WU Jing-wei, LIN Jia01 Insitute of Command and Control Engineering,Army Engineering University of PLA,Nanjing 210007,China<br/>2 The Army Artillery and Defense Academy of PLA,Nanjing 211100,China<br/>3 The Key Laboratory of Polarization Imaging Detection Technology,Hefei 230031,China<br/>4 Shandong Military Region,Ji'nan 250000,ChinaSince the camouflage object is highly similar to the surrounding environment with a rather small size,the general detection algorithm is not fully applicable to the camouflaged object detection task,which makes the detection of camouflaged object more challenging than the general detection task.In order to solve this problem,the existing methods are analyzed in this paper and a detection algorithm for camouflage object is proposed based on the YOLO v5 algorithm.A new feature extraction network combined with attention mechanism is designed to highlight the feature information of the camouflage target.The original path aggregation network is improved so that the high,middle and lowly level feature map information is fully fused.The semantic information of the target is strengthened by nonlinear pool module,and the detection feature map size is increased to improve the detection recall rate of the small size target.On a public camouflage target dataset,the proposed algorithm is tested with 7 algorithms.The mAP of the proposed algorithm is 4.4% higher than that of the original algorithm,while the recall rate has improved 2.8%,which verifies the effectiveness of the algorithm for camouflaged object detection and the great advantage in accuracy compared with other algorithms.<br/>http://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2021-10-226.pdfcamouflaged object|object detection|attention mechanism|yolo|aggregation network |
| spellingShingle | WANG Yang, CAO Tie-yong, YANG Ji-bin, ZHENG Yun-fei, FANG Zheng, DENG Xiao-tong, WU Jing-wei, LIN Jia Camouflaged Object Detection Based on Improved YOLO v5 Algorithm Jisuanji kexue camouflaged object|object detection|attention mechanism|yolo|aggregation network |
| title | Camouflaged Object Detection Based on Improved YOLO v5 Algorithm |
| title_full | Camouflaged Object Detection Based on Improved YOLO v5 Algorithm |
| title_fullStr | Camouflaged Object Detection Based on Improved YOLO v5 Algorithm |
| title_full_unstemmed | Camouflaged Object Detection Based on Improved YOLO v5 Algorithm |
| title_short | Camouflaged Object Detection Based on Improved YOLO v5 Algorithm |
| title_sort | camouflaged object detection based on improved yolo v5 algorithm |
| topic | camouflaged object|object detection|attention mechanism|yolo|aggregation network |
| url | http://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2021-10-226.pdf |
| work_keys_str_mv | AT wangyangcaotieyongyangjibinzhengyunfeifangzhengdengxiaotongwujingweilinjia camouflagedobjectdetectionbasedonimprovedyolov5algorithm |