Segmentation of Moving Vehicles under Complex Outdoor Condition Using Block Matching Algorithm
P. Richard Rajkumar, A. Rajalakshmi, B Janani, Robert Bosch · 2014
In the presence of unfavorable luminance conditions or in the presence of camera noise, the segmentation and motion detection of traffic vehicles in an outdoor environment, particularly under non ideal weather condition such as snowfall, heavy rain, fog is still an area of active research. In computer vision system, to detect moving objects Gaussian-based background modeling is used. But it has some limitation. So we propose an approach using the block matching algorithm, which robustly detects the motion of interest and can suppress the false motion in a challenging outdoor environment. To measure in BMA, The sum of absolute difference (SAD) is commonly used and this determines the motion/change in video sequences. BMA was adopted by many video-coding standards such as MPEG- 1/2/4, H.261, H.263, and H.264/AVC. FS or exhaustive search algorithms consider every pixel in the block and provide optimum results This algorithm segment multiple vehicles and even able to detect the human intervention in an unrestricted area.