Application Of Mathematical Morphology And Fractal Theory In Image Denoising And Edge Detection

Image denoising can improve image quality and signal-to-noise ratio effectively, get more information carried by the original image, lay a good foundation for the digital image processing. The edge of the image contains a lot of image information, it is an important step in image processing to detect the image edge accurately, image edge detection has been a hotspot in the research of image processing.This article mainly discusses the image denoising and edge detection with combinations of mathematical morphology and fractal theory, and it is applied to process remote sensing images.This article introduces the development history of image denoising and present research status, analysises the difficulties in image denoising. According to the characteristics of morphology and fractal theory, morphological opening-and-closing operation can solve this problem effectively. After morphological filtering, using the fractal theory to wipe off low frequency part of noise, also can play the role of smoothing image to remove the burr produced by direction restriction elements of morphology.Morphological gradient operator can resist the interference of noise and detect edge position accurately. But the edge contain a lot of texture, aiming at this shortcoming, using the fixed window to divided the edge, calculate the fractal dimension of each window. Depending on the fractal dimension value to distinguish the edge from the image.The proposed algorithm is applied to process remote sensing image. The experimental results show that this algorithm can obtain high signal-to-noise ratio while removing the noise and detect edge from remote sensing image accurately, it can restrain the interference of texture effectively.