Abnormal behavior identification method and system based on human body key points

• Main Authors: WANG XIAOHU, LIU SHAOZHENG, DU HAIYAN, KONG BIGUANG, LIU FANBO, QIN JIANMING, SONG DENGFENG, JIN CHUNXIAN, FENG JIANHUI, CHEN LE, ZHANG YUMEI, HE CHUNYUAN, XU ZHENGGUO, LI SHENG, HE ZHENGMEI, LEE JIN HONG, PU BICAI, YING ZICAI, JIANG XIAOJING, DAI LIQIN, YANG XIANGJUAN, SHE YOUMING, SU HUI, LI JUNHONG, HUO ZHIFENG, ZHAO DAN, GENG ZUOXUE, MA WENLIANG, ZHANG TIEBIN, YANG YONGSEN, LU JIANFENG, YANG YONGPING, LIU ZHEN, ZHAO ZHENGPING, SONG MINGMING, ZHANG YING, GENG HUAIXU, DANG LULU, YANG ZHENGHONG, YANG JIE
• Format: Patent
• Language: Chinese; English
• Published: 15.09.2023
• Subjects: ALARM SYSTEMS; CALCULATING; COMPUTING; COUNTING; ORDER TELEGRAPHS; PHYSICS; SIGNALLING; SIGNALLING OR CALLING SYSTEMS

The invention discloses an abnormal behavior recognition method and system based on human body key points, and belongs to the technical field of high-altitude operation monitoring. Using a key point detection algorithm to extract position information of joint points in the aerial work picture, and obtaining postures and actions of the operator in the aerial work picture and postures and actions of the operator in the aerial work picture; if the body of the operator in the aerial work picture exceeds the electronic fence and reaches a set body area threshold value, judging that the operator exceeds a safety range, and triggering an alarm prompt; and judging whether the operator has an abnormal behavior in the aerial work picture, and if the operator has the abnormal behavior in the aerial work picture, triggering an alarm prompt to notify the operator to stop working. The safety state of operators is monitored in real time, an alarm is given in time, and the safety of high-voltage line electric power operation