Accelerating Reconstruction of Reflective Fourier Ptychographic Microscopy by Employing a Global Optimal Search Algorithm in a Graphics Processing Unit

Abstract

Reflective-type Fourier ptychographic microscopy, an outstanding solution for surface inspection, employs a phase retrieval algorithm to reconstruct a high-resolution image with a wide field-of-view. This technique is time-consuming because it requires processing a large number of low-resolution images coming from the dark-field illuminator and bright-field illuminator. To decrease the computation time, we propose a new Fourier ptychography recovery method running in parallel in a graphics processing unit. We also utilized an adaptive step-size incremental gradient descent searching algorithm. The approach was validated by measuring a USAF reflective resolution target and by inspecting the surface of a smartphone integrated circuit. The processing time was remarkably reduced to 88.29 s, about 11.1 times less than the state-of-the-art method, while obtaining about 250 nm half-pitch resolution and about 1 × 1 mm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> field-of-view.