Local Chaotic Encryption Based on Privacy Protection on Video Surveillance
Suolan Liu, Lizhi Kong · Proceedings of the 2018 8th International Conference on Social science and Education Research (SSER 2018) · 2018
Privacy protection has been a hot research topic in recent years with the widely application of surveillance systems.Most developed mechanisms focused on modifying the monitored targets partly or wholly in the surveillance scenes.These methods are usually irreversible of privacy protection and have negative influences on the subsequent action recognition in intelligent monitoring system.In this paper, a novel privacy preserving scheme is produced by only encrypting human face region.To properly localize human face, we firstly do pedestrian detection and then extract face region by using skin color information in YCbCr color space, which is the predominant difference between face and other parts of a person.Furthermore, an improved spatial chaotic map is developed to encrypt face region.The performance evaluation on a video clip with complex background shows that the proposed scheme can effectively and robustly track pedestrian and obscure the face region in real time.