A Compact Multi-frame Blind Deconvolution Algorithm With The Fourier Constraints

When observing the spatial objective of the satellite,the space shuttle and the space station by using the large ground-based optical telescope,the observation images will be severely blurred due to the presence of atmospheric turbulence,observation noise and the Optical system aberration,which seriously affects the recognition and monitoring of the space target.The multi-frame blind deconvolution algorithm is a common method of image post-processing.The MFBD algorithm uses a series of short exposure images,which is estimated to improve the quality of adaptive optical images in a series of physical constraints.However,it is a pathological question,and in the solution,it is necessary to bind the algorithm with the physical characteristics of the target image,the physical process of imaging and the physical characteristics of the noise.In order to solve the problems of poor noise suppression effect and slow restoration speed of MFBD algorithm,this paper studies the compact multi-frame blind deconvolution algorithm based on frequency domain constraints: Firstly CAOS simulation software is used to generate a large number of short exposure degraded images,the spectrum mask of each degraded image is calculated by using the frequency domain constraint,and a image-selection method is proposed by using the spectral mask method,in the method a certain percentage of the high quality image is selected as “control frame” which is directly involved in image restoration,the rest of the image is regarded as “non-control frame” which only plays a role of supplying constraints,Finally,the programming of compact blind deconvolution algorithm is realized.The performance of the algorithm is verified by processing spatial objects,and the restoration effect and speed of the algorithm are improved.On the basis of the implementation of the compact blind deconvolution algorithm,the high-precision target image estimated by the compact blind deconvolution algorithm is used as the initial value for further image restoration processing,and the improved compact blind deconvolution algorithm is completed,which further improves the restoration effect of the algorithm compared with the compact blind deconvolution algorithm.