Moving Object Detection in Stereo Video Sequences

Moving object detection starting from a video sequence is a fundamental task in many computer vision applications, such as autonomous robotics, traffic control, driver assistance and surveillance systems. In this paper we propose a method, based on stereo vision, that improves separation ability. Two image sequences of the same scene are taken using two cameras with slight horizontal displacement. Stereo processing of these sequences gives both detection of movement and depth information, which can be used to distinguish moving objects with different distance from the observer. Information on the distance of a given point of the image can be useful in presence of moving objects appearing close in a 2D image but with different distances from the observer.