Abstract
Multiple distance phase retrieval methods hold great promise for imaging and measurement due to their less expensive and compact setup. As one of their implementations, the amplitude-phase retrieval algorithm (APR) can achieve stable and high-accuracy reconstruction. However, it suffers from the slow convergence and the stagnant issue. Here we propose an iterative modality named as weighted feedback to solve this problem. With the plug-ins of single and double feedback, two augmented approaches, i.e. the APRSF and APRDF algorithms, are demonstrated to increase the convergence speed with a factor of two and three in experiments. Furthermore, the APRDF algorithm can extend the multiple distance phase retrieval to the partially coherent illumination and enhance the imaging contrast of both amplitude and phase, which actually relaxes the light source requirement. Thus the weighted feedback enables a fast-converging and high-contrast imaging scheme for the iterative phase retrieval.
Highlights
Phase retrieval methods1, which have been applied in various fields of science and engineering, including X-ray imaging2, quantum imaging3, astronomy4, super-resolution5, diffraction tomography6, and wide-field imaging7, enable one to recover the phase information of an object from the magnitude of its diffraction pattern
A fiber laser with the wavelength of 532 nm is used for the coherent illumination and a target of ‘HIT’ etched on a glass is regarded as the sample
To quantitative show this improvement, the logarithm of mean square error (LMSE) curves are plotted in Fig. 6(p), where the acceleration effect is remarkable for the weighted feedback
Summary
Phase retrieval methods, which have been applied in various fields of science and engineering, including X-ray imaging, quantum imaging, astronomy, super-resolution, diffraction tomography, and wide-field imaging, enable one to recover the phase information of an object from the magnitude of its diffraction pattern. The original GS algorithm was slow and sensitive to initial guesses To solve this problem, Fienup improved the GS algorithm with a support constraint of non-negativity and a pre-assigned boundary to speed up the convergence under single or double measurements. Fienup improved the GS algorithm with a support constraint of non-negativity and a pre-assigned boundary to speed up the convergence under single or double measurements9,10 These methods both have to get a rough estimation for the object. Without the need of the support constraint, these methods are capable of obtaining an optimal convergence and high-accuracy reconstruction by utilizing different scanning strategies, such as overlapping illumination, multi-wavelength scanning, multi-angle illumination, pinhole scanning and multiple distance measurements. The result indicates that the weighted feedback APR is capable of obtaining an enhanced imaging contrast for different objects
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
