New Operator for Image Edge Detection Based on Finite Difference Method

Detection of edges in an image is a very important step towards understanding the image features and it is considered as the most common approach for detecting meaningful discontinuities in gray level. The edges in an image are detected using the derivative of the image. The method that uses the first derivative is called gradient while the one that uses the second derivative is called Laplacian. This paper is concerned in the edge detection problem using the first derivative method. Our proposed method of edge detection is based on the implicit finite difference, where we have only used the first derivative of the images for x and y directions in order to construct the two masks. The proposed masks are more general than the classical gradient masks like Sobel, Prewitt, and Robert. Our proposed method has been implemented and tested on artificial images, and then compared with Sobel and Prewitt edge detectors using a performance measure. The results showed good detection of edges in low contrast images in comparison with Sobel and Prewitt edge detectors.