Phase Diversity Wave-front Sensing Technique And Its Application In The Imaging Of Space Objects

The phase diversity wave-front sensing technique(PD)is a phase recovery wave-front sensing technique,which extracts information of the amplitude and phase from at least two images of the same target,with a condition that the two main images have a known phase diversity.In contrast,PD does not rely on atmospheric statistical characteristics,nor does it require reference target.The optical imaging system for PD also is simple and only requires a little additional hardware,thus reducing the error introduced into the system.Besides,each photon in PD can be used for normal imaging and estimating the wave-front errors.Wave-front calculations are independent of image content and are suitable for point target and extended target.These advantages of PD make it possible to monitor the wave-front state of space cameras in orbit,and provide powerful technical support for space cameras to detect space targets.Based on the technical advantages of PD,this paper carries out theoretical simulation,algorithm improvement and outfield experimental verification.The research content and research results of this thesis mainly are as follows:1.Adaptive analysis was carried out for the sensitivity of PD to five main factors.Based on Matlab,the cost function of PD under ideal condition and noise condition are completed respectively,and regular terms are introduced into the cost function to improve the convergence speed of the PD.Then,the BFGS optimization algorithm is written to achieve high-precision wave-front restoration and high quality image reconstruction.Subsequently,the five main fac tors affecting the applicability of PD are simulated and analyzed.The results show that the introduced defocus scale should be controlled from 0.2? to 1.2?,the defocus control precision should be controlled in 0.15? to 0.75?,and the system wave-front distortion should be in the range of 0.3? to 1.2?,Gaussian noise's variance should be controlled from 0.0001 to 0.0005,and the size of the reconstructed image should be smaller than 128?128 pixels.Through the above adaptive analysis of PD's sensitivity to five main factors,it provides effective parameter support for practical experiments and engineering applications.2.Wave-front coding technique(WFC)is applied to achieve effective extension of PD's applicability.In order to enhance PD's applicability,the wave-front distortion allowed by PD is effectively extended by WFC,which is insensitive to the accuracy of defocus control,thereby enhancing the applicability of PD.The results show that the adaptive range of PD to the defocus scale can be effectively extended with the help of the cubic square mask,and the adaptive range of the defocus scale is extended by nearly 8 times.Similarly,the range of PD's adaptability to wave-front distortion has nearly doubled.Through the above analysis,the application of WFC provides effective technical support for the expansion of the adaptability range of PD.3.When PD detects the space target,due to the external factors like long-distance and low illumination,the image reconstruction process is susceptible to severe interference from Poisson noise.Therefore,the NLM algorithm is introduced as a pre-processing process,and the noise-prevented image is first denoised.The obtained image is then brought into the established PD model for image reconstruction,thereby minimizing the existence of Poisson noise,ensuring that the fine structure in the captured image is no t damaged,and achieving high quality image reconstruction.The effectiveness of the modified PD technique is verified by numerical simulation and field experiments.The analysis results show that compared with the traditional PD,the wave-front RMS of the modified PD decreases by nearly 51% in the range of Poisson noise from 24.48 d B to 61.30 d B.Compared with the traditional PD,the overall decline of the SSIM value of the modified PD is reduced from 47% to 17%.The results of the Ordos outfield experiment and numerical simulation analysis show the same improvement trend,and then fully prove the practical performance of the modified PD for space object reconstruction.