A robust line extraction method by unsupervised line clustering

This paper describes a new method of extracting straight Lines based on uns upervised line clustering. It is assumed that each line support region (LSR ) in an image is composed of pixels that share similar gradient orientation values. Therefore, by an appropriate partitioning of gradient space, the s ets of parallel lines can be more easily extracted. Previous works on parti tioning gradient space, however, relied on ad hoc methods, and cannot be us ed as reliable tools for the extraction of the number of clusters in gradie nt space. In order to handle such a clustering issue, the Bhattacharyya dis tance is introduced to define a measure for cluster separability and therea fter to estimate the number of inherent clusters. Subsequent to the cluster ing stage, each extracted line support region undergoes a consistency test to evaluate its validity In terms of uncertainty descriptors. For the consi stency test, an entropy-based line selection scheme is formulated and a the ory from robust statistics is adopted. The feasibility of the proposed line extraction method is assessed by considering the issue of vanishing point detection. (C) 1999 Pattern Recognition Society. Published by Elsevier Scie nce Ltd. All rights reserved.