Quantitative imaging for complex-objects via a single-pixel detector

Abstract: Quantitative phase imaging (QPI) is important in many applications such as microscopy and crystallography. To quantitatively reveal phase information, people could either employ interference to map phase distribution into intensity fringes, or analyze intensity-only diffraction patterns through phase retrieval algorithms. Traditionally, both of these two ways use pixelated detectors. In this work, a novel QPI scheme is reported inspired by single-pixel camera (SPC), which adopts the principle of SPC that retrieves images through structured illumination and corresponding single-pixel signals. Particularly for complex-valued imaging, the structured illumination is performed in the phase domain, and a point detector with restricted sensor size detects the intensity of zero-frequency area. Based on the illumination structures and point signals, a complex image is reconstructed by running a phase retrieval algorithm. This approach is universal for various wavelengths, and needs no a priori information of the targets. Both simulation and experiment show that our single-pixel QPI scheme exhibits great performance even with objects in an extremely rough phase distribution.