Measuring and Comparison of Edge Detectors in Color Spaces

Edge detection has been a popular practice in image processing and computer vision applications. Many image processing applications require the discovery of edge details in the gray or color images as a beginning stage of an image processing, vision and understanding. Generally, edge detection on grayscale images is not affluent enough to explain intensity changes. Therefore, we can use color edge information as an important method. Because the result is different when input images are color images or not (grayscale images). The main purpose of the proposed edge detection is to discern significant parts from the normal features in a given image. We assume that intensity varies rapidly in a significant part. There are many color spaces such as RGB, YIQ, and HSV (Hue, Saturation, and Value). In this paper, we conducted edge detection on each color spaces and compared the results. Simulation results show that the HSV color space gives the best detection performance.