Pendeteksian Tepi Citra Digital Berwarna Menggunakan Sobel Mask Yang Ditingkatkan dan Median Filter

Abstract

Edge detection is a fundamentally important thing to do in its application to digital images. This is because edge detection can produce the edges of an image in a short time. One of the methods commonly used in edge detection is the Sobel Mask. However, Sobel Mask only convoluted in horizontal and vertical directions. Besides that, Sobel Mask is also sensitive to noise. Therefore, the Sobel Mask is improved by adding a new kernel mask and combined with the Median filter. The new kernel mask is obtained by rotating the horizontal and vertical kernel masks by 45 degrees. In addition, edge detection is carried out on RGB images which are generally carried out on grayscale images. The improved Sobel Mask provides more detailed edge detection with firmer lines compared to the regular Sobel Mask.