Matching polygons using canonical string representation

Document Type

Conference Proceeding

Publication Date

12-1-2007

Abstract

In image retrieval applications, one of the boundary-dependent approaches is matching contours with their polygonal representation. In the previous study [14], a novel efficient polygonal shape representation that gives a unique representation was introduced. The representation is invariant under similarity transformation and it can be utilized for affine distortion as well. In this study, robustness of the representation is explored and complexity of the algorithm is proved as 0(n log n), and two algorithms to match polygons with different number of vertices is proposed. The first algorithm has an exponential computation complexity but it is optimal, whereas the later has a polynomial complexity, but it may not give the optimal solution.

Publication Source (Journal or Book title)

Proceedings of the 2007 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2007

First Page

407

Last Page

413

This document is currently not available here.

Share

COinS