안전한 도시철도 환경을 위한 방치, 제거된 물체 검출 알고리듬의 구현

초록

In this paper, a detection system for the abandoned or removed objects is proposed and implemented using DSP-based embedded system for urban transit safety. To detect those objects, we proposed an object detection system using GMM(Gaussian mixture model) and image histogram in ROI(region of interest) on the input images. Our system is optimized for fast processing in embedded system. Experiments showed that the proposed system works with high accuracy in various environments.