Abstract: Image edge is the most basic feature of image. The edge is the set of pixels which has step change in pixel gray value. Image edge reflects most of the image information. Therefore, edge detection is an important part of image processing. The review has shown that the still much improvements can be done in the edge detection. Most of the classical edge detection methods take operation on the neighbor region pixels, and obtain the gradient with templates approximation, such as Robert, Sobel, and Prewitt, which are relatively simple and easy to implement, and have good real-time performance, but these operators are sensitive to noise, poor anti-interference performance. In order to overcome the limitations of the earlier work a new approach has been proposed for color images using L*A*B color model, color gradients, particle swarm optimization based improved canny edge detector i.e. L*A*B based Canny. The L*A*B color space has ability to efficiently reflect the difference in human eye and color sensation. Color based edge detection has 10 % more potential edges than the gray one. The particle swarm optimization based edge detection can successfully reduce the poor speed issue with ant colony optimization. The color based gradients has the ability to remove the effect of the false edges while preserving the potential edges.
Keywords: Pixels; Detection; Particle swarm optimization; canny edge detector; Gradient.