Phase Diversity Technique And Its Improvements In Faint Object Reconstruction

The phase diversity(PD)wavefront sensing technique is used to simultaneously infer phase aberrations and reconstruct an object from two or more degraded images of the same object.A known phase diversity is introduced between these images.The phase diversity technique has developed to have an important role in adaptive and active optic systems.A deep theoretical and experimental study has been operated based on the major applications of PD technique in wavefront sensing and image reconstruction.Firstly,the fundamental principles of an incoherent imaging system are analyzed in spatial and frequency domain: the OTF is the normalized autocorrelation function of the generalized pupil function;and the influences of the aberrations on the OTF.The basic principles of phase diversity wavefront sensing technique are fully introduced;the objective function and the expression for the object are derived by using maximum-likelihood estimation based on Gaussian noise model.Secondly,the objective function is optimized by a proper optimization algorithm to get the phase aberrations and the pristine object.Various optimization algorithms have been presented,such as Newton method,steepest descent method,Powell method,and particle swarm optimization.The results obtained by steepest descend method have indicated that it cannot meet the criterion in some conditions.In numerical simulations,0.1?~ 0.5? root-mean-square(RMS)random phase aberrations are used to investigate the PD technique from images degraded by these random phase aberrations.Root-mean-square errors(RMSE)of phase aberrations detected by particle swarm optimization is less than 72 10 ?-? or 4 105%-?.Particle swarm optimization is effective for PD technique as a post image processing technique.Thirdly,in order to study phase diversity(PD)technique quantitatively,experiments has been performed.An experimental system based on Liquid Crystal Spatial Light Modulator(LC-SLM)is established,and the problem of lacking quantitative data for atmospheric turbulence phase screen can be solved effectively.The feasible and practicability of PD technique is verified by experiment results.Finally,essential information buried in Poisson noise reduces the estimation precision of phase aberrations and deteriorates the quality of reconstructed images for PD technique,while observing a faint object in astronomic domain.In order to solve this problem,an effective method is proposed.The denoising algorithm based on block-matching and 3D filtering is introduced in the wavefront sensing field as a preprocessing stage.Then,the PD technique is applied to the denoising images.The RMSEs of phase estimates on synthetic data are decreased by approximately 40% across noise levels within the range of 58.7-18.8 d B in terms of peak signal-to-noise ratio(PSNR).Meanwhile,the overall decline range of SSIM is significantly decreased from 49% to 9%.The experiment and simulation results are in good agreement.Results of the numerical simulations and experiments demonstrate that our approach is effective for reducing the sensitivity of PD technique to Poisson noise.