COMPARISON AND APPLICATION OF SOBEL AND CANNY FILTERS

5 5

Abstract

The Sobel operator is a gradient method used in edge detection. It determines the change in intensity (gradient) in the image, and also calculates changes in the horizontal (x) and vertical (y) directions of the image. And Canny edge detection is an advanced way of finding edges. Both methods are used to detect edges, but each has advantages and disadvantages. It follows several steps and aims to detect stable edges with minimal errors. This study aims to compare the performance of Sobel and Canny edge detection algorithms on different image datasets. Although the Sobel algorithm gave faster results in terms of processing time, the Canny algorithm showed superior performance in terms of accuracy and noise tolerance. In tests on natural, medical, and industrial images, the Canny algorithm provided higher accuracy and noise tolerance compared to Sobel. In terms of processing time, the Sobel algorithm gives faster results. The research results show that the Canny algorithm should be preferred in applications that require high accuracy and noise tolerance, and the Sobel algorithm in cases where fast processing time is important. These results provide key information to help select an appropriate edge detection algorithm in various application scenarios.