Cellular Automata for Edge Detection Based on Twenty-Five Cells Neighborhood

Cellular Automata is a complex system that has been widely and successfully utilized in image processing to handle tasks such as edge detection, noise filtering, enhancement, smoothing, feature selection, thinning, convex hulls, and so on. A novel edge detection approach based on Cellular Automata is provided in this study. To cope with the challenge of edge detection, an extended Moore neighborhood is investigated. The proposed edge detector is evaluated on a variety of images. The resulting findings are compared to those obtained using the Canny, Sobel, Laplacian, and Scharr edge detection techniques. The quality of the produced edges is measured using fitness functions such as RMSE and SSIM. In addition, the execution time is compared. Experiments show that the proposed strategy produces excellent outcomes.