Mathematical Morphology Edge Detection Algorithm of Remote Sensing Image with High Resolution

Based on characteristics of remote sensing images, with high spatial resolutions and mathematical morphology (MM) methods, MM edge detection algorithm were developed to process remote sensing images using multi-scale omni-directional structure elements. In this algorithm, the peak signal-to-noise ratio (PSNR) technique was created to acquire the self-adaptive scale weights and replace the method with fixed mean. Results suggest that: the purposed algorithms, compared with the traditional Canny edge detection operator and MM edge detection algorithm with simplex scale figure structure elements, can successfully resolve the contradictions between noise suppression and extraction of fine edge excellently, and anti-noise performance is strong. The image edge information can be used in extraction geometrical and texture features.