Multiple human body tracking using the fusion of CCD and thermal image sensor

摘要

Human body detection and tracking is an important research field of computer vision in monitoring applications of the healthcare, surveillance and etc. The performance of detection and tracking is closely associated with each other. In this paper, a novel approach for detection and tracking a moving body in field of sensor fusion with respect to CCD, thermal image sensor is proposed. We calibrated CCD and thermal image by using the chess board which can display thermal and texture data for heterogeneous sensor fusion. The Segmented Blob Tree of body was developed to track the moving bodies. Experiments showed that the method yields good detection and tracking performance in dynamic environments, such as a lighting change, various electric heating appliances or complicated backgrounds.