Design and Implementation of an Integrated ...

In this study, a new technique using Shannon entropy and based on the fuzzy logic is used for the edge detection from aerial/satellite images. In the proposed technique, different information layers are determined by the input image based on different values of threshold and using Shannon entropy, and then the appropriate edges are automatically detected and extracted using the combination of information layers. In order to evaluate the algorithm, the results from the proposed technique are compared with that from LOG, Sobel, Prewitt edge detectors and a fuzzy edge detector technique. The results show that the hybrid system presents higher reliability for the detection of image brightness and contrast variations, curve-shaped features and sharp corners.