Fast detecting moving objects in moving background using ORB feature matching

Moving objects detection in moving background is a prerequisite for many analyses and applications in computer vision. The most challenging part of moving objects detection in moving background is that motion induced by camera moving may dominate the observed motion, which makes existing methods cannot detect the real moving objects from the moving background robustly and computationally efficient. In this paper, we proposed a very fast method that can detect moving objects from moving background accurately without prior knowledge of the camera motion characteristics. Frame difference that caused by camera motion is compensated by Oriented FAST and Rotated BRIEF (ORB) features matching. Mismatched features between two frames are rejected by the proposed method for a good accuracy of compensation. To validate the proposed algorithm, we test it on three real videos and compare its performance with Scale Invariant Feature Transform (SIFT) based and Speeded-Up Robust Features (SURF) based method. The results show that the proposed method is at forty times faster than SIFT based method and twenty times faster than SURF based method, while detecting more accurately in many situations.