In order to build a safe and reliable public security intelligent monitoring platform, it is studied from 3 aspects: authentication, tracking and identification. The intelligent monitoring platform for public safety is constructed by using the methods of face bio-assay, finger vein recognition, long-term pedestrian tracking, human behavior analysis and legacy detection, and using the distributed system architecture. The application results show that the application improves the robustness, accuracy and real-time of identity authentication, intelligent video surveillance target tracking and behavior analysis, and legacy detection in public places.