An Obstacle Segmentation Approach in ALV System

In ALV system, obstacle detection algorithm is one of the most important research fields and obstacle segmentation is a key factor in obstacle detection approaches. In this paper, a simple and fast segmentation approach is proposed for ALV system based on binocular stereo vision. Firstly, a simple method is applied to binarize the images by the defined binarization function; Secondly, in order to save the computation time of gray mean of pixels in squared window, the method of reduce the redundant computation is applied in our algorithm; Lastly, in order to avoid the affection of inner region of obstacle objects, the outer contour tracing algorithm is applied to get the outer contour of likely obstacle in binarized images, as a result, the likely obstacle objects are segmented. Several images captured from ALV system is used to test the proposed algorithm, the result indicate that the approach is valid and feasible.