Motion object detection method based on real-time background update under complex environment

Abstract

For the fast requirement of motion object detection under complex environment, a background subtraction motion object detection method based on real-time background update is presented in this paper. Firstly the color images distance between the current frame and background is calculated. And the state variable is defined to describe the objects motion change state in the current frame image in order to update the real-time background image. Then the motion object detection can be completed by the background subtraction result between real-time background and the current frame image. At last, mathematical morphology operation can be used to eliminate the isolated noise and motion objects can be detected quickly. This method can adapt to all kinds of complex mutation. Actual color images from traffic surveillance system are used to test the performance. The experimental result verifies its accuracy and effectiveness of the proposed algorithm.