Orientation Space Filtering for Multiple Orientation Line Segmentation

概要

論文の詳細を見る
The goal of this paper is to present an appropriate line segmentation method for intersections (X-junctions) and branches (T-junctions). There are multiple orientations in local regions of the intersections and branches. A novel representation called "orientation space" is proposed, which is derived by adding the orientation axis to the abscissa and the ordinate of the image. The orientation space representation is constructed by treating the orientation parameter, to which the Gabor filters can be tuned, as a continuous variable. The problem of multiple orientation line segmentation is dealt with by thresholding 3D images of the orientation space and then detecting the connected components therein. In this way, X-junctions and T-junctions are able to be separated effectively. We show experimental results using synthesized images and real biomedical images. Especially in x-ray coronary angiogram, the overlapped vessels were segmented by orientation space filtering.
社団法人電子情報通信学会の論文
1998-02-19