期刊文献+

一种巡逻执勤目标检测算法研究

Research on a Patrol Duty Object Detection Algorithm
下载PDF
导出
摘要 巡逻执勤是具有重要意义的安全维稳行动,但是巡逻环境复杂、目标多样、检测难度大的问题十分突出,所以如何准确、实时检测巡逻执勤目标具有重大现实意义。为了提升对巡逻执勤目标检测的准确性和实时性,基于YOLOv5算法进行改进。为抑制巡逻环境带来的干扰,结合ECANet注意力机制进行改进,提高被检测目标显著性;同时为保证较好的实时性及多尺度目标检测能力,引入BiFPN网络结构。将改进算法与原始算法进行比较,mAP提升3.51%;与4种算法进行了对比实验,结果显示该算法能较好地降低巡逻执勤目标检测因检测相似、尺度多样、光照干扰等问题带来的影响,进一步验证了该算法在巡逻执勤目标检测任务中的有效性。 Patrol duty is a security and stability maintenance operation of great significance, but the patrol environment is complex, the object are diverse, and the problem of difficult detection is very prominent, so how to accurately and real-time detect patrol duty objects is of great practical significance. In order to improve the accuracy and real-time detection of patrol duty objects, the YOLOv5 algorithm is improved. In order to suppress the interference caused by the patrol environment, the ECA-Net attention mechanism is combined to improve the saliency of the detected object;and the introduction of BiFPN structure ensures better real-time performance and multi-scale object detection capabilities of the algorithm. Comparing the improved algorithm with the original algorithm, the mAP is improved by 3. 51%;comparing with four algorithms, the results show that the algorithm can better reduce the impact of patrol object detection of due to the problems of similar detection,diverse scales and light interference, which further verifies the effectiveness of the proposed algorithm in the task of patrol duty object detection.
作者 岳磊 袁建虎 杨柳 吕婷婷 YUE Lei;YUAN Jianhu;YANG Liu;LÜTingting(Field Engineering College,Army Engineering University of the PLA,Nanjing 210001,China)
出处 《现代防御技术》 北大核心 2023年第1期67-74,共8页 Modern Defence Technology
关键词 目标检测 巡逻执勤 注意力机制 BiFPN 深度学习 计算机视觉 object detection patrol duty attention mechanism BiFPN(bi-directional feature pyramid network) deep learning computer vision
  • 相关文献

参考文献8

二级参考文献72

  • 1李明,鹿朋,朱美强,姜瑾,邹亮.基于改进YOLO-tiny的闸板阀开度检测[J].煤炭学报,2021,46(S02):1180-1190. 被引量:5
  • 2刘浩,刘海滨,孙宇,王竞陶,黄辉.煤矿井下员工不安全行为智能识别系统[J].煤炭学报,2021,46(S02):1159-1169. 被引量:24
  • 3杨东华,李久贤,卞治国.Marr边缘检测算法的研究[J].中国图象图形学报,2006,11(6):823-826. 被引量:17
  • 4江志军,易华蓉.一种基于图像金字塔光流的特征跟踪方法[J].武汉大学学报(信息科学版),2007,32(8):680-683. 被引量:34
  • 5Yilmaz A,Javed O,Shah M.Object tracking:a survey[J].ACM Journal of Computing Surveys,2006,38(4):1-45.
  • 6Sun Y,Yuan B.Hierarchical GMM to handle sharp changes in moving object detection[J].Electronics Letters,2004,40(13):801-802.
  • 7Yin Jianqin,Han Yanbin,Hou Wendi,et al.Detection of the mobile object with camouflage color based on optical flow[J].Procedia Engineering,2011,15:2201-2205.
  • 8侯文迪,李建,李金屏.基于局部微分光流与拓扑一致性的伪装色移动目标检测[C]//第16届中国图象图形会议论文集,2012:630-634.
  • 9Li Jinping,Li Qijuan,Yin Jianqin.Real-time tracking by combining level set and meanshift[J].Journal of Information&Computational Science,2008,5(2):829-836.
  • 10Wang Zhaowen,Yang Xiaokang,Xu Yi.Camshift guided particle filter for visual tracking[J].Pattern Recognition Letters,2009,30:407-413.

共引文献178

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部