Abstract: Aiming at the problem of safety helmet wearing detection on engineering project site, based on the target detection algorithm, this paper proposes a front and rear end separation method to realize the intelligent detection and display of safety helmet wearing. The back-end mainly uses the YOLO V3 algorithm as the core to detect and analyze the live camera video and pictures, and store the violation data into the database. The front-end uses the web development framework to build an intelligent algorithm platform to visually display the illegal data in the database. The algorithm and implementing platform have been applied in the actual project site, and the analysis results show that the YOLO V3 algorithm model has high calculation accuracy and real-time detection speed.
