Photon-Limited Imaging by Timestamped All-Pixel Sensing (PITAS)

Abstract

Single-photon technology, based on the measurement of very weak-dose electromagnetic radiation fields, has diverse applications ranging from life sciences to remote sensing. Meanwhile, computational algorithms based on such sparse photons speed up the reconstruction and optimization tasks in low-light-level imaging. However, conventional photon-counting strategies leave out the time parameter, potentially improving image quality, which has been an open but unresolved challenge. Here, we propose photon-limited imaging by time-stamped all-pixel sensing to retrieve the objects in low-light conditions. By simultaneously timestamping the observed photons in all pixels, effectively extracting the intertime information, and globally optimizing the correlation of adjacent pixels, we experimentally achieve a 2.25-fold enhancement in contrast superior to the photon counting method with the same limited photons. The proposed scheme is not limited by the light sources, the kernel of which may be of considerable value to weak pump and probe processes in the fields of fluorescence microscopy, biomedical imaging, particle physics, astronomy, etc.