OPTICAL FLOW BASED MOTION DETECTION USING BIOLOGICALLY MOTIVATED SELECTIVE ATTENTION MODEL

摘要

This paper proposes a new motion detection method using the optical flow in a selective attention model for object tracking. The selective attention model focuses on the interested regions of an image as human visual system does. The proposed motion detection method uses the selective attention model results for identifying an interesting object and tracks the target using the optical flow information when the identified object moves. The optical flow for object tracking is obtained by the saliency map information instead of using intensity information of an image, which is effective for detecting not only a moving object with a fixed camera but also a moving object with a moving camera. Experimental results show that the proposed method gives the outstanding performance for detecting a moving object even when the light intensity is changed.