Research On Image Restoration In Diffraction Telescope Imaging

To meet the urgent need for high-resolution imaging in space application fields such as cosmic exploration and earth observation,the space telescope aperture needs continuously increasing.However,the traditional large-aperture reflecting space telescope' main lens has strict requirements on surface accuracy,difficulty in folding,and the increasing weight and volume of the system,which seriously restricts its application in space field.In comparison,the diffractive telescope' main lens is similar to a plane and easy to fold.The surface manufacturing requirement is much lower.The weight and volume of the system are significantly reduced.Therefore,the diffractive space telescope has become one of the important technical routes for large-aperture space telescopes with a diameter of more than 10 meters.The multi-order diffraction,however,results to the image contrast of the diffractive telescope system low and susceptible to noise.For extended object imaging,how to improve the diffraction telescope image contrast and suppress noise is an urgent problem.To solve the above key technical issues,this paper establishes a theoretical model of imaging degradation combining the imaging characteristics of diffraction telescopes,and then using image restoration technology conducts related theoretical and experimental studies,to improve diffraction telescopes high-contrast clear imaging.The following works have been done:First,the simplified imaging model of the diffraction imaging system is derived by the Fresnel approximation formula and the Fresnel diffraction lens transmittance function.The generalized pupil function has been revised to include not only the limitation of the aperture shape and the system traditional aberration,but the effect of the special aberration non-designed diffracted light.It indicates that the imaging degradation caused by the multi-order diffraction phenomenon can be characterized by the degradation function,which provides a concise theoretical calculation method for the imaging performance analysis of the diffraction imaging system,and provides theoretical guidance for the study of image restoration algorithms suitable for diffraction telescopes.Then the research on non-blind deconvolution restoration algorithms of the diffraction telescope is carried out based on the imaging degradation characteristics of the diffractive telescope.An adaptive Wiener filtering(Awn)algorithm and an adaptive block matching 3D collaborative filtering image restoration(ABM3D)algorithm are proposed.Numerical simulation and experimental research are carried out relying on prototype imaging platforms of diffraction telescope systems.The consistent results of simulation and experiment indicate that the Awn algorithm can effectively improve the image contrast,the algorithm structure is simple,the calculation speed is fast,but the image noise is amplified.It is suitable for scene imaging with low noise level.ABM3 D algorithm is anti-noise strong and has high image contrast recovery ability at the cost of computational efficiency.It is suitable for post-processing of diffraction imaging system images.The research of non-blind deconvolution restoration algorithm provides new technical ideas for high-contrast and clear imaging of diffraction telescopes.Based on the research of the non-blind deconvolution restoration algorithm,further research on the effect of the acquisition method of the degradation function for the diffraction telescope image restoration is carried out.The degradation function can be obtained by optical design software simulation,calculation based on the simplified imaging model,or measurement in the laboratory.The accuracy of the degradation function directly affects the image restoration accuracy.Based on three degradation functions,two algorithms proposed in this paper were used to restore the diffraction telescope system images.Simulation analysis and experimental results show that the non-blind deconvolution image restoration of the diffractive telescope system should choose the actually measured degradation function.And then,based on the laboratory measurement values of the degradation function,the two algorithms were used to restore the real scene images obtained by the diffraction telescope prototype.Comparing the real scene images with and without restoration,the sharpness of the restored image is significantly improved.Aiming at the image quality degradation problem caused by multi-order diffraction of the diffractive telescope system,this paper establishes a theoretical model of imaging degradation,proposes effective image restoration algorithms,and clarifies the optimal measurement method of the degradation function.It provides the theoretical basis and technical guidance for the subsequent research on suppressing the non-design order diffracted light in the diffraction telescope system by image restoration technology,and lays the foundation for the high resolution and high contrast imaging of the diffraction telescope system.