Curb segments detection with temporal filtering for urban driving scenarios

Abstract

An improved approach for the detection of curbs (border of relevant traffic isles, sidewalks, etc) is presented, in the context of urban driving assistance systems. A rectangular elevation map is built from 3D dense stereo data. Edge detection is applied to the elevation map in order to highlight height variations. Temporal filtering is performed for edge points, based on the ego car motion, and only persistent points are validated. Relevant lines are extracted using a Hough accumulator. Curb segments are extracted from each relevant line by applying a set of proposed conditions. The algorithm was evaluated in an urban scenario. It works in real-time and provides robust detection of curbs.