Research on recognition of safety helmet wearing of electric power construction personnel based on artificial intelligence technology

Abstract

The traditional safety helmet wearing recognition is only based on the color, shape, texture and other characteristics of the image, which is greatly affected by the external environment, and has the problem of unstable recognition accuracy. In view of the above problems, this paper studies the recognition method of power construction workers' safety helmet based on artificial intelligence technology. After preprocessing the construction monitoring image, such as graying and denoising, the construction personnel in the detection image are located, that is, on the basis of detecting the construction personnel area, the head position of the construction personnel is located, and finally the safety helmet wearing recognition is realized by using YOLO algorithm. The simulation results show that the average recognition accuracy is 95.2%, the recognition effect is stable and has good robustness.