Research on the application of OpenPose in escalator safety systems

Abstract

Abstract This paper describes the application of the OpenPose based human pose detection model in escalator safety systems. The YOLO target detection is used to obtain the position of the human body in the image, and then the human limb information is extracted by OpenPose, and the human joint connection algorithm connects the identified human limb information to obtain the human pose information in the image. The experimental results show that the algorithm can effectively identify the human body posture and can provide timely alarms for dangerous postures, thus improving the safety of escalator riders.