安全、健康和环境2024，Vol.24Issue(12)：1-8,8.DOI:10.3969/j.issn.1672-7932.2024.12.001
基于机器视觉的危化品事故场景人因行为检测研究进展
Research Progress on Human Behavior Detection in Hazardous Chemical Accident Scenes Based on Machine Vision
摘要
Abstract
Accidents in the hazardous chemical indus-try are characterized by the high flammability and ex-plosiveness of the materials involved,as well as the high temperature and pressure of the production process.When an accident occurs,the timely and ef-fective emergency response largely determines the subsequent development of the accident.Human e-mergency response behavior is crucial in the rescue process.In recent years,a large number of machine vision based methods for identifying human factors in hazardous chemical accident scenarios have been pro-posed.The existing machine vision based methods for detecting human behavior in hazardous chemical scenes was sorted out,and the experimental dataset of human behavior detection in hazardous chemical scenes was summarized to provide new references for subsequent related research.The relevant experimen-tal datasets were introduced,and the performance of existing methods through experiments was compared and analyzed.Firstly,the human behavior detection methods for hazardous chemical scenes based on ma-chine vision in recent years were introduced.Then the advantages and disadvantages of different methods were comprehensively summarized,and their applica-ble scenarios were analyzed.Subsequently,statistical analysis was conducted on the existing dataset of hu-man behavior detection in hazardous chemical acci-dent scenarios,and the detection dataset was classi-fied according to data types.The problems and im-provement methods in the existing dataset were pro-posed.Finally,the development trend of human be-havior detection in hazardous chemical accident sce-narios based on machine vision was discussed.关键词
危化品事故/智能视频监控系统/人因行为/机器视觉/行为识别/行为检测方法Key words
hazardous chemical accidents/intelligent video surveillance system/human behavior/machine vision/behavior recognition/behavior detection methods分类
资源环境引用本文复制引用
王春..基于机器视觉的危化品事故场景人因行为检测研究进展[J].安全、健康和环境,2024,24(12):1-8,8.基金项目
国家重点研发计划(2023YFC3011100),面向大型复杂火场的举高喷射消防机器人关键技术研究及应用. （2023YFC3011100）
