Performance Analysis For Adaptive Optics System And Restoration Method Investigation For Turbulence Degraded Image

The random fluctuation of the atmospheric refractive index caused by atmospheric turbulence can cause the distortion of the wavefront,seriously affect the imaging performance of the optical system and reduce the resolution of the observed object.Adaptive optics technology is one of the most effective way to overcome the atmospheric turbulence,but due to the influence of the adaptive optics system performance and noise,the wavefront error can be partially corrected,the observed target image resolution is still far from the diffraction limit.Therefore,the subsequent restoration processing is critical to the turbulence degraded image.In order to improve the resolution of the turbulence image,this paper focuses on the research of turbulence degraded image denoising and restoration algorithm:First,research on adaptive optics system.Introduced structure,working and imaging principle of the adaptive optics system,analyzed parameters and performance of the 61 unit adaptive optical system in yunnan observatory 1.2 m telescope,and designed light path diagramof a double corrector adaptive optics system.After the analysis,it is known that the correction of the wavefront aberration of the adaptive optics system is not sufficient,the follow-up processing of the observed turbulence images is necessary.The second,research on wavefront reconstruction technology and turbulence wavefront phase screen simulation.Studied of Zernike polynomial wavefront reconstruction technology,by using the Zernike polynomial method to simulate the turbulent atmospheric turbulence phase screen distribution under different terms.From the simulation results can be seen,when Zernike polynomial number is lower,turbulence screen lack of high frequency components,and the low frequency component is outstanding,phase stationary.With the increase of Zernike number of polynomials,the turbulence distribution on the turbulence screen gradually slow,decline phase stability.Once again,research on denoising algorithm for turbulence degraded image based on wavelet domain Curvelet transform.The idea of multiscale geometric transform is introduced to the image denoising.Firstly,the principle of Curvelet transform in wavelet domain is introduced.And then the adaptive threshold selection method is improved according to Bayes criterion,extraction of high-frequency coefficients of two-dimensional wavelet transform and taking Curvelet transform based on Wrapping on it.Finally,achieved Cuevelet transform based on wavelet domain of turbulence degraded image denoising processing.The experiment results show that the algorithm has a clear visual effect,PSNR and MSE are significantly improved that compared with the original image.The last,research on restoration algorithm for atmospheric turbulence image based on improved maximum likelihood estimation.Firstly,the turbulence image is denoised.And then the Zernike polynomial is used to restore the turbulence degraded image,the PSF initial value is estimated based on the wavefront information.The ML algorithm is improved by combining the prior information of the image,introducing the target edge preserving constraint and the PSF constraint.In the iterative process of target image estimation and PSF estimation,the bandwidth constraint of PSF is performed.Finally,the algorithm is used to recover the simulated turbulence image and the real turbulence image.The experiment results show that the proposed algorithm can effectively improve the resolution of the reconstructed image,and the gray mean gradient of the restored image is improved obviously,image energy is more concentrated.