Space Target Image Super-resolution Reconstruction Study

As more and more platforms are sent to the outer space, people pay much more attention to the space object images. It is obviously a good method to build the space object images according to the space object imaging results by means of the Super-Resolution reconstruct. The reconstructed image is approximate to or even in excess of the light diffraction limited ones.Focus on the research of the space image Super-Resolution , the works done in this thesis are presented as following:1 The fundamental principle of Super-Resolution reconstruction is analyzed and for the first time, the probability of Super-Resolution reconstruction in intuitionistic method is interpreted.2 After presenting the image models, characters and existing methods of blind image deconvolution, Iterative Blind Deconvolution(IBD) algorithm Maximum Likelihood (ML) algorithm, Richard-Lucy (RL) algorithm, Nonnegativity and Support Constrants Recursive Inverse Filtering (NAS-RIF) algorithm are analyzed and realized, then based on a lot of experiments the algorithms are compared with each other.3 A new method using Maximum Likelihood to deal with the speckle image is presented. This method can reconstruct extended objects from several noised speckle frames without reference star data.4 The blind method of defining image support region is studied and object support region is defined through image threshold, which obtains good result.5 Based on the classical median filtering, an adaptive and weighting Filter method is proposed which works well in experiments.