The Target Image Based On The Total Variation Minimization Space Recovery

The recognition of space object is a key step in space defence. Using space object images with super-resolution to obtain their geometrical form is necessary to observe and catalog space object, predict their future position and determine their orbits. Because of optical distortion induced by atmosphere turbulence and limitation of optical devices, acquired images of space object are blurred and degraded. It's an economic scheme for us to improve image resolution by using quick and effective post-processing method. Focus on the super-resolution restoration of space object, research and experiment are performed, which consists of the following parts:1. The imaging model and regularization methods of the inverse problem in imaging processing are analyzed.2. Super-resolution algorithm based on Total Variation minimization is researched and programming performed.3. Muti-grid algorithm is applied on space object restoration, which reduced the runtime of TV algorithm.4. APEX algorithm is analyzed and realized on computer which obtained an ideal reconstruction result.5. IBD algorithm is analyzed, and the space IBD method combined with the IBD method based on Fourier transformation is realized. For improving the ill-posed problem of the space IBD method, an adaptive operator is proposed in this paper. The results show that the improved algorithm works well.