Abstract: Edge detection plays an important role in image processing, pattern recognition and in computer vision applications. The amount of data to be processed is reduced in image edge detection while preserving important structural properties in an image. Various edge detection operators such as Sobel operator, Prewitt operator, Robert operator has been proposed in the literature for extracting the edges from the grayscale image. Only 90% of edge information in a color image can be found in the corresponding grayscale image. It implies that 10% of the edges are left over in grayscale images. Since color images give more information than grayscale images, this 10% left over edges may be extracted from color images. In the proposed algorithm, edge at each pixel of an image is calculated using fuzzy rules around 3 * 3 spatial masks. Fuzzy inference system designed has 8 inputs, which corresponds to 8 neighboring pixels of instantaneous scanning matrix, one output that tells whether the pixel under consideration is an edge or not. Rule base comprises of thirty rules, which classifies the target pixel. The proposed method results for both color images and gray scale images.The fuzzy inference system allows effective utilization and representation of human expertise about the system, thus the edges can be detected efficiently.

Keywords: Edge detection, Color edge detection, Fuzzy logic, Fuzzy inference system.