Kernel sets in compass edge detection

Abstract

Edge detection is an important operation in the analysis of a digital image and includes a set of mathematical methods to identify points of discontinuity. For this purpose, different algorithms are used for estimating image gradient. Many edge detection techniques are implemented with convolution mask and based on discrete approximations to differential operators. Differential operations measure the rate of change in the image brightness function. The compass approach implies taking a single mask and rotating it to major compass orientations. Using a suitable programming environment, this paper provides a comparative visual analysis of several digital images, with the ability to change the kernel set in compass edge detection. © 2013 IEEE.