Zebra-crossing detection for the partially sighted

Zebra-crossings are useful road features for outdoor navigation in mobility aids for the partially sighted. In this paper, zebra-crossings are detected by looking for groups of concurrent lines, edges are then partitioned using intensity variation information. In order to tackle the ambiguity of the detection algorithm in distinguishing zebra-crossings and stair-cases, pose information is sought. Three methods are developed to estimate the pose: homography search approach using an a priori model; finding normal using the vanishing line computed from equally-spaced lines and with two vanishing points. These algorithms have been applied to real images with promising results and they are also useful in some other shape from texture applications.