Computer Engineering and Applications ›› 2023, Vol. 59 ›› Issue (7): 242-249.DOI: 10.3778/j.issn.1002-8331.2208-0061

• Graphics and Image Processing • Previous Articles     Next Articles

Optimizing Human Abnormal Behavior Detection Method of YOLO Network

ZHANG Hongmin, ZHAUNG Xu, ZHENG Jingtian, FANG Xiaobing   

  1. 1.School of Electrical and Electronic Engineering, Chongqing University of Technology, Chongqing 400054, China
    2.Liangjiang International College, Chongqing University of Technology, Chongqing 401135, China
  • Online:2023-04-01 Published:2023-04-01

优化YOLO网络的人体异常行为检测方法

张红民,庄旭,郑敬添,房晓冰   

  1. 1.重庆理工大学 电气与电子工程学院,重庆 400054
    2.重庆理工大学 两江国际学院,重庆 401135

Abstract: Because of the large interference of environmental background information in public surveillance videos and the different scale of abnormal human behavior goals, at present, it is difficult to improve the precision of human abnormal behavior detection. For the above issues, this paper designs the abnormal behavior detection method by improving the YOLOv5 module. In this method, a shielded convolutional attention model is added to the original YOLOv5 backbone network. The module starts from a shielded convolutional layer, and the central region of the receptive field is covered. The shielding information is predicted and the errors related to the shielding information are used as abnormal scores. At the same time, Swin-CA module is embedded in the detection network. Through the study of characteristics of adjacent layers, enables the module to get stronger grasp the overall situation information, thus reducing the affect of backdrop message on the detection results, by extracting the scale characteristics of human behavior abnormalities in different backgrounds, it decreases the order of complex of the whole model calculation and improves the precision of the module to locate the target of abnormal human behavior. Experimental results on the UCSD-PED1, KTH and Shanghai Tech datasets show that the precision of the proposed method reaches 98.2%, 96.4% and 95.8%, respectively.

Key words: abnormal human behavior, YOLOv5, mask convolution, attentional mechanism, Swin-CA module

摘要: 鉴于公共场合监测视频信息中周围环境背景信息干扰大以及人体异常行为目标的尺度不同,目前人体异常行为检测的准确性难以进一步提高。针对上述问题,设计了通过改进YOLOv5网络的异常行为检测方法。该方法在原YOLOv5主干网络添加屏蔽卷积注意力模型,该模块从一个屏蔽卷积层开始,感受野的中心区域被遮掩,通过预测屏蔽信息并利用与屏蔽信息相关的误差作为异常得分。在检测网络中嵌入Swin-CA模块。通过对相邻层特征的学习,使得模型能够更好地掌握全局信息,从而减小了背景信息对检测结果的影响,通过提取不同背景中人体异常行为尺度特征,降低了整个模型计算的复杂度,提高了模型对人体异常行为目标定位的精度。在UCSD-ped1、KTH和Shanghai Tech数据集上的实验结果表明,提出方法的检测精度分别达到了98.2%、96.4%和95.8%。

关键词: 人体异常行为, YOLOv5, 屏蔽卷积, 注意力机制, Swin-CA模块