Color Image Edge Detection: A Survey
International Journal of Innovations in Engineering and Technology · 2017
This paper presents a comprehensive review on various color edge detection techniques.Several edge detection techniques have been proposed in literature to detect edges present in monochrome as well as color images.Some commonly used first derivative edge operators and second derivative edge operators such as Roberts, Sobel, Prewitt, Compass, Laplacian of Gaussian (LoG), Canny, Marr-Hildreth and Haralick are discussed here and then the techniques are further extended to vector-valued techniques.In the traditional color edge detection techniques, the color image is first separated into three different channels such as R, G and B before processing and then the algorithm is applied on the respective channels and in the end all channels are combined to produce the result.But in the Quaternion based techniques the color pixel is considered as a single unit as a result of which the geometrical transformations and three dimensional rotation are easier to perform on it compared to the RGB domain.