Based On Wavelet And Spirit Removed

Cloud cover noise often occurs in remote sense image processing. It is a very important task to remove the effects of cloud cover. How to realize it is the necessary stage of bind restoration of remote sense image processing.This paper introduces a way to remove the effects of cloud by frequency transform. Here, information of cloud cover is removed by high-pass filtering and information of background is enhanced by homomorphic filtering and spatial-frequency analysis. Furthermore, a new way is particularly discussed based on wavelet transform.After strict tests, we find that multiresolution analysis based on wavelet transform can improve the effect of the method of Fourier transform. The results show this approach is effective when there is massive cloud cover on the remote sense image.Wavelet analysis is internationally recognized up to the minute tools for analyzing time-frequency. It is chiefly due to the "adaptive feature" and "mathematical micro-telescope feature". This paper introduce the application of wavelet to image processing.This paper has such special points as:1. Based on the idea of removing cloud cover, a new mathematical model has been constructed. This paper introduces a way to improve the effect of removing cloud cover by spatial-frequency analysis and a new method to improve the effect of removing cloud cover on color image by IHS transform.2. Wavelet transform and conventional high-pass filtering method have been combined. Using multiresolution analysis of wavelet transform, details of the result image after being removed cloud cover are greatly improved.