中文

Journal of Applied Sciences ›› 2021, Vol. 39 ›› Issue (4): 605-614.doi: 10.3969/j.issn.0255-8297.2021.04.008

• Special Issue on CCF NCCA 2020 • Previous Articles    

Recognition Method of Human Dangerous Behavior in Multimodal Scenes Using Reinforcement Learning

ZHANG Xiaolong1, WANG Qingwei2, LI Shangbin3   

  1. 1. P. E. Department, Northeast Forestry University, Harbin 150040, Heilongjiang, China;
    2. Physical Education Department, Harbin Huade University, Harbin 150025, Heilongjiang, China;
    3. Physical Education Department, Harbin Engineering University, Harbin 150001, Heilongjiang, China
  • Received:2020-08-30 Published:2021-08-04

PDF

555

Abstract: In multimodal scenes, conventional human dangerous behavior recognition methods generally perform low recognition accuracy. Therefore, this paper proposes a human dangerous behavior recognition method based on reinforcement learning. Firstly, a feature extraction algorithm of reinforcement learning is used to obtain feature subsets of human dangerous behavior in multimodal scenes. Secondly, human dangerous behaviors in multimodal scenes are extracted by reinforcement learning data decision-making, and a fuzzy recognition model of human dangerous behavior is constructed. Finally, by bringing the obtained feature subsets of human dangerous behavior into the model and calculating the membership degree of dangerous behavior under different senses, the recognition of human dangerous behavior in multimodal scenes can be realized. Experimental results show that the method in this paper has a high recognition accuracy and a recognition delay of less than 300 ms.

Key words: reinforcement learning, multimodality, scene, behavior recognition

CLC Number: