Object detection based on sparse representation model for power safety supervision

Abstract

In order to detect whether power operators wear safety equipment, a sparse representation model based target recognition algorithm is proposed for power intelligent safety monitoring in this paper. In the face of hard hats, protective clothing and other target detection, the current target detection algorithm can not guarantee the detection accuracy. The sparse representation model is used to enhance the super-resolution of the target image under low resolution, and then the enhanced image is detected, and finally the behavior of wearing safety helmet and goggles is determined, so as to meet the new demand of power intelligent safety supervision. The experimental results show that the proposed algorithm has higher detection accuracy than the existing methods and can meet the requirements of power monitoring system.