A Comparison of Sobel and Prewitt Edge Detection Operators

Abstract

This paper presents a comparative analysis of Sobel and Prewitt edge detection operators, focusing on their performance in detecting edges in digital images. Both methods are widely utilized in image processing to enhance and identify object boundaries. The study employs the Berkeley Segmentation Dataset 500 (BSDS500) for evaluation and MATLAB for implementation. Key performance metrics such as accuracy and execution time are analyzed. Results indicate that while Prewitt outperforms Sobel in terms of edge clarity and precision, Sobel shows a slight advantage in computational efficiency. This research provides valuable insights into the selection of edge detection techniques for various image processing applications