Adaptive and efficient Fourier ptychographic microscopy based on information entropy

Abstract

Fourier ptychographic microscopy is a simple and effective computational imaging method that improves the spatial bandwidth product of the system. Its principle is to realize the aperture expansion by using a phase retrieval algorithm to splice a set of low-resolution images under different illumination angles in the Fourier domain. However, the long periods of time required for the data acquisition limit its application. Therefore, we propose a new adaptive FP algorithm to significantly reduce the volume of data acquired using the information entropy of the image as a judgment function, and a special separation strategy as a judgment method to estimate the information distribution of a high-resolution sample from a specific low-resolution image. Further, the effectiveness and superiority of our method are verified via simulations and experiments. The results show that the volume of data acquired can be reduced by more than 60% without any deterioration in resolution and quality of the recovered image.