Corner Detection to Capture Significant Two Dimensional Intensity Changes

Neeta Nain, Vijay Laxmi, and Nemi C. Singh

Keywords

Corner, curvature, chain code, consistency

Abstract

A corner is a 2D discontinuity corresponding to a point in an image where there are significant intensity variations. This paper presents a new approach for robust, scale independent corner detection using difference chain codes as curvature to capture significant intensity changes in two dimensions. Simplicity and efficiency of the technique is inherent owing to use of only integer operations. Evaluation with respect to Harris and CSS corner detectors has been done and consistency, accuracy and repeatability of the method is verified using the existing techniques.

Important Links:



Go Back