Research On Edge Detection Of High Resolution Remote Sensing Images By Zernike Moments

In recent years, satellite remote sensing has made rapid development, the resolution remote sensing images in the continuous improvement, large numbers of high resolution remote sensing images used in remote sensing makes continuously expanding. But at the same time, the difficulty of remote sensing image processing has become larger and larger. Traditional remote sensing image processing this information only for the spectral image, and can not reflect the characteristics of the entire remote sensing, especially high-resolution remote sensing images, it has more features (spectral, texture, geometric features, etc.). Edge is an important geometric features of remote sensing image feature extraction for remote sensing images of buildings, roads and so has a very important role. Edge detection technology is the core technology of extraction of remote sensing image edges. In addition, edge detection or remote sensing image processing, to provide basic data services. Therefore, edge detection for remote sensing image processing has become one of the hot current study remote sensing. In this paper, the characteristics of remote sensing images, combined with Zernike moments high spatial resolution remote sensing image edge detection. The main contents of this paper include the following aspects:1, on the present situation of research and edge detection techniques were summarized and analyzed, elaborated classic gray image edge detection method, and a relatively new technology in these theories formed on the basis. Looking edge detection trends in the application of remote sensing image scores.2, research Zernike moment method, and detailed application of this method in a digital image edge detection in the analysis of this method has drawbacks.3, for the Zernike moments in high remote sensing image edge detection in the presence of the problem, remote sensing image edge detection method score a combination of FCM Zernike Moments and Zernike Moments presents a combined score of thresholding remote sensing image edge detection method.4, in order to verify the accuracy of two kinds of edge detection method set forth above, we use the two comparative experiments. The method of single Zernike moment the first group were raised edge detection result obtained by comparison; a second group with the proposed method with the classical edge detection operator for comparison. By comparing the results of visual analysis, the proposed method is an effective score of remote sensing image edge detection method.5, analyze and summarize the results of this paper achieved and problems, and the prospect of further work.