Research On Remote Sensing Image Processing Techniques Based On Compressed Sensing

In conventional Nyquist theory,only the sampling frequency reaches twice of the signal bandwidth,the signal can be accurately reconstructed.However,with the explosive development of information,it will lead to huge amounts of data to be processed in the collection of images or video according to the conventional sampling theory.Thus,it greatly increases the costs of hardware and software,and the most important is that the efficiency of this method is very low.In conclusion,the conventional Nyquist sampling method for large amount of data to process fall down.In order to break up the restrictions,in recent years,a new theory of sampling-Compressed Sensing(CS)is presented by Candes and Donoho et.It brings vitality into the field of information processing,and expand the research direction.CS sample has many advantages:first of all,it greatly reduces the amount of data to be processed;secondly,it can accurately reconstruct the original signal far below the Nyquist sampling times.This paper studies the application of remote sensing image processing based on Compressed Sensing:On account of large amounts of data needed process in image fusion,this paper proposes a remote sensing image fusion method with compressed sensing based on wavelet sparse basis.The first step is to get I,H,S components of multispectral image,then match the panchromatic image with I component,the next step is conducting weighted fusion in CS domain,and then reconstruct new I component and conduct IHS inverse transformation of the fused image.Experiments show that this algorithm can get good quality of fusion image,more important is the reduction of data processed because of fusion in CS domain.An adaptive total variation de-noising algorithm based on Compressed Sensing is proposed in this paper,and it can de-noise pointedly because of selecting different global variable factors according to different region characteristics of image.Experiments show that the algorithm can not only remove noise of remote sensing image,but also protect image edge information,the most important is that it can reduce the amount of data needed for processing.