IMPROVEMENT OF BACKGROUND UPDATE METHOD FOR IMAGE DETECTOR
Y-J Soh Kim · 1998
Image-based vehicle detection has been playing an important role in traffic surveillance. There are several methods for detecting vehicles in the image. One of them is background differencing method. Background differencing method requires two steps: background extraction and background update. The most frequently used method for background update is filtering. Since, in the method, updating is done based on pixels in the same location in a series of frames, it does not adequately handle passive shadow caused by stationary structures in road scene. In this paper, an improved update scheme based on local neighbor of pixel of interest is proposed. For the covering abstract see IRRD E102946.