A novel edge detection algorithm for remote sensing images based on the self-similarity of fractal character

A novel edge detection algorithm for remote sensing images, which combines the image gray level gradient and the fractal self-similarity character of image edges based on fractal theory, is introduced. The self-similarity coefficient between the element block and the local block, which are both centered at the current pixel being processed, in addition to gray level gradient decision-making criteria, are used to make the decision about the existence of image edges. A binary operator is then used to threshold its magnitude and produce the edge map of the image. The results of the experiment are presented which show the effectiveness and the noise resistance of the proposed new algorithm for remote sensing image edge detection.