Analysis of Local Image Structure using Intersections of Conics
We propose an algorithm for the analysis of local image structure that is able to distinguish between a number of different structures like corners, crossings, y-junctions, t-junctions, lines and line segments. Furthermore, parameters of the detected structures can be evaluated, as, for example, the opening angle of corners. The main idea of the algorithm is to fit the intersection of two conics to the local image structure. The results of the algorithm when applied to synthetic and real data will be presented.
Use and reproduction:
No license. The provisions of the German Copyright Act (UrhG) apply.
Please note that individual components of the publication may be subject to other licensing or copyright conditions.