Research on a Patrol Duty Object Detection Algorithm

doi:10.3969/j.issn.1009-086x.2023.01.009

Abstract

Abstract:

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.

Key words: object detection, patrol duty, attention mechanism, BiFPN（bi-directional feature pyramid network）, deep learning, computer vision

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ECA-Net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。

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

CLC Number:

E919

Lei YUE, Jianhu YUAN, Liu YANG, Tingting LÜ. Research on a Patrol Duty Object Detection Algorithm[J]. Modern Defense Technology, 2023, 51(1): 67-74.

岳磊, 袁建虎, 杨柳, 吕婷婷. 一种巡逻执勤目标检测算法研究[J]. 现代防御技术, 2023, 51(1): 67-74.

Figures/Tables 11

Fig. 1 Network partition flow chart of the YOLOv5

Fig. 2 Focus schematic of the process

Fig. 3 Improved detection algorithm framework

Fig. 4 Efficient channel attention mechanism module

Fig. 5 Add channel attention mechanism effect contrast

Fig. 6 Three feature network structure

Table 1 Experimental environment configuration

配置名称	版本参数
GPU	NVIDIA Quadro P6000
CPU	Intel（R） Xeon（R） Gold 5118CPU@2.30GHz×2
操作系统	Windows 10
CUDA	11.5

Fig. 7 mAP@0.5 and mAP@0.5：0.95 training change curves

Table 2 Comparison of ablation experiment results

Data Enhancements	ECA-Net	BiFPN	mAP @0.5：0.95/%	mAP @0.5/%	t/s
			36.03	73.67	0.013
√			37.43	74.71	0.021
	√		37.95	75.74	0.031
√	√		38.78	76.64	0.035
√	√	√	39.12	77.18	0.038

Table 3 Comparative experiments with each detection network

Name	mAP@0.5%	t/s
YOLOv4	70.12	0.089
YOLOv5	73.67	0.013
SSD	68.36	0.235
Faster R-CNN	78.05	2.855
Ours	77.18	0.038

Fig. 8 Comparison of various detection algorithms

References 20

1	侯文迪，王宁，李金屏. 基于运动信息与Hog特征的伪装色移动目标跟踪［J］. 计算机工程与应用， 2014， 50（24）： 187-190.
	HOU Wendi， WANG Ning， LI Jinping. Camouflaged Moving Object Tracking Algorithm Based on Motion Information and Hog Feature［J］. Computer Engineering and Applications， 2014， 50（24）： 187-190.
2	苏昂. 无人机图像中的典型目标检测与变化检测方法研究［D］. 长沙：国防科学技术大学， 2016.
	SU Ang. Research on Typical Object Detection and Change Detection from Unmanned Aerial Vehicle Images［D］. Changsha： National University of Defense Technology， 2016.
3	南晓虎，丁雷. 深度学习的典型目标检测算法综述［J］. 计算机应用研究， 2020， 37（S2）： 15-21.
	Xiaohu NAN， DING Lei. Review of Typical Object get Detection Algorithms for Deep Learning［J］. Application Research of Computers， 2020， 37（S2）： 15-21.
4	史彩娟，张卫明，陈厚儒，等. 基于深度学习的显著性目标检测综述［J］. 计算机科学与探索， 2021， 15（2）： 219-232.
	SHI Caijuan， ZHANG Weiming， CHEN Houru， et al. Survey of Salient Object Detection Based on Deep Learning［J］. Journal of Frontiers of Computer Science and Technology， 2021， 15（2）： 219-232.
5	刘星，陈坚，杨东方，等. 尺度约束辅助的空对地目标智能检测方法［J］. 现代防御技术， 2019， 47（2）： 73-79.
	LIU Xing， CHEN Jian， YANG Dongfang， et al. Scale-Constrained Air-to-Ground Object Intelligent Detection Method［J］. Modern Defence Technology， 2019， 47（2）： 73-79.
6	GIRSHICK R. Fast R-CNN［C］∥2015 IEEE International Conference on Computer Vision （ICCV）. Santiago： IEEE，2015： 1440-1448.
7	REN Shaoqing， HE Kaiming， Ross Girshick， et al. Faster R-CNN： Towards Real-Time Object Detection with Region Proposal Networks［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence，2017，39（6）： 1137-1149.
8	REDMON J， DLVVALA S， GIRSHICK R， et al. You Only Look Once： Unified， Real-Time Object Detection［C］∥2016 IEEE Conference on Computer vision and Pattern Recognition （CVPR）. IEEE， 2016： 779-788.
9	REDMON J， FARHADI A. YOLO1000： Better， Faster， Stronger ［C］∥2017 IEEE Conferonca on Computer Vision and Pattern Recognition （CVPR），Honlulu， HI， USA： IEEE， 2017： 6517-6525.
10	WEI L， ANGUELOV D， ERHAN D， et al. SSD： Single Shot Multibox Detector［C］//Proceedings of the 14th European Conference. Netherlands： Springer， 2016： 21-37.
11	赵晓枫，徐明扬，王聃漂，等. 基于改进SSD的特种车辆红外伪装检测方法［J］. 红外与激光工程， 2019， 48（11）： 116-125.
	ZHAO Xiaofeng， XU Mingyang， WANG Danpiao， et al. Infrared Camouflage Detection Method for Special Vehicles Based on Improved SSD［J］. Infrared and Laser Engineering， 2019， 48（11）： 116-125.
12	邓小桐，曹铁勇，方正，等. 改进RetinaNet的伪装人员检测方法研究［J］. 计算机工程与应用， 2021， 57（5）： 190-196.
	DENG Xiaotong， CAO Tieyong， FANG Zheng . et al. Research on Detection of People with Camouflage Pattern via Improving RetinaNet［J］. Computer Engineering and Applications， 2021， 57（5）： 190-196.
13	SONG Tainian， QIN Weiwei， LIANG Zhuo， et al. Research on CNN for Anti-missile Object Detection Algorithm Based on Improved Attention Mechanism［C］∥The 40th Chinese Control Conference （CCC 2021）， 2021： 23-28.
14	WANG C Y， LIAO H Y M， WU Y H，et al ．CSPNet：A New Backbone that can Enhance Learning Capability of CNN［C］∥2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops（CVPRW）.Seattle，USA， 2020：1571-1580．
15	HE Kaiming， ZHANG Xiangyu， REN Shaoqing， et al. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition［J］∥IEEE Transactions on Pattern Analysis and Machine Intelligence，2015， 37（9）： 1904-1916.
16	SHU Liu， LU Qi， QIN Haifang， et al. Path Aggregation Network for Instance Segmentation［C］∥2018 IEEE/CVF Conference on Computer Vision and Pattern Recogition. Salt Lake City， UT， USA， 2018：8759-8768.
17	WANG Qilong， WU Banggu， ZHU Pengfei， et al. ECA-Net： Efficient Channel Attention for Deep Convolutional Neural Networks［C］∥2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seatle， WA， US： IEEE， 2020：11531-11539.
18	郝帅，张旭，马旭，等. 基于CBAM-YOLOv5的煤矿输送带异物检测［J］. 煤炭学报， 2022， 47（11）： 4147-4156.
	HAO Shuai， ZHANG Xu， MA Xu， et al. Foreign Object Detection in Coal Mine Conveyor Belt Based on CBAM-YOLOv5［J］. Journal of China Coal Society， 2022， 47（11）： 4147-4156.
19	TAN M， PANG R， LE QV. EfficientDet： Scalable and Efficient Object Detection［C］∥2020 IEEE/CVF Computer Society Conference on Computer Vision and Pattern Recognition.Seattle， WA， USA， 2020：10778-10787.
20	赖秋频，杨军，谭本东，等. 基于YOLOv2网络的绝缘子自动识别与缺陷诊断模型［J］. 中国电力， 2019， 52（7）： 31-39.
	LAI Qiupin， YANG Jun， TAN Bendong， et al. An Automatic Recognition and Defect Diagnosis Model of Transmission Line Insulator Based on YOLOv2 Network［J］. Electric Power， 2019， 52（7）： 31-39.

[1]	Yuchen SHI. Review of Underwater Target Wake Detection in Remote Sensing [J]. Modern Defense Technology, 2024, 52(1): 83-91.
[2]	Mengxue KANG, Jinpeng ZHANG, Zhe MA, Xuhui HUANG, Yating LIU, Zizhuang SONG. High-Level Semantic Distillation for Incremental and Continuous Object Detection [J]. Modern Defense Technology, 2024, 52(1): 41-48.
[3]	Yang LI, Xiao LIANG, Haisa JIA. Ship Detection in SAR Images Based on Knowledge Distillation [J]. Modern Defense Technology, 2023, 51(4): 78-85.
[4]	Zhe LI, Yiqi TONG, Wenbo XIA, Yuxin YING. A Multi-source Centralized Air Target Classification Method [J]. Modern Defense Technology, 2023, 51(4): 53-62.
[5]	Ming HE, Zihan ZHU, Xulong ZHAI, Zheng ZHAI, Chengpeng HAO. Air-to-Ground Target Detection Algorithm Based on Multi-branch Context Fusion [J]. Modern Defense Technology, 2023, 51(3): 91-98.
[6]	Baoyin YAO, Lei MAO, Zhibin WANG. Applications of Artificial Intelligence in Space Equipment [J]. Modern Defense Technology, 2023, 51(2): 33-42.
[7]	Jia-wei BAO, Xiao-ping TIAN, Yu-na LIU, Chang-kuan ZOU, Yu-qing ZHANG. Summary of Research on Security Issues of ADS-B Protocol [J]. Modern Defense Technology, 2022, 50(5): 28-35.
[8]	YAO Bao-yin, MAO Lei, XIAO Ke, QU Hui. Applications of Artificial Intelligence on the Optical Earth Observation [J]. Modern Defense Technology, 2021, 49(5): 26-31.
[9]	ZHOU Li, SHI Jian, ZHANG Qing-bo, CHEN Guang-yang, YANG Shuang-ling. Research on Physical and Functional Damage Assessment of Phased Array Antenna [J]. Modern Defense Technology, 2021, 49(4): 56-63.
[10]	WANG Wen-qing, PANG Ying, LIU Yang, YANG Dong-fang, ZHANG Meng. Air-to-Ground Target Intelligent Detection Algorithm Based on Dual Attention Mechanism [J]. Modern Defense Technology, 2020, 48(6): 81-88.
[11]	WANG Zhi-hu, SHEN Xiao-qing, GUI Wei-long. Survey of Small Targets Detection Technology in Optical Imaging [J]. Modern Defense Technology, 2020, 48(5): 67-73.
[12]	TANG Run-ze, ZHANG Cheng-long, LI Lin-lin. Application of Artificial Intelligence on Situation Assessment and Game Countermeasure in Unmanned Battlefield [J]. Modern Defense Technology, 2020, 48(5): 25-31.
[13]	MA Xin-xing, TENG Ke-nan, HOU Xue-long. Forecast Model of Main Air Attack Direction Based on CNN [J]. Modern Defense Technology, 2018, 46(5): 6-12.