Bionic retina-like first-photon imaging

First-photon imaging is a photon-efficient imaging technique that reconstructs an image by using only the first arrival photon event at each spatial location. However, the image reconstruction depends on the spatial correlations of natural scenes, which causes degradation of image quality in low-resolution first-photon imaging. First-photon imager with high-resolution sampling has a long data acquisition time due to the low probability of detecting a single photon. Inspired by the biological structure of the human's retinas, we proposed a space-variant resolution sampling structure with a high-resolution foveal region and low-resolution peripheral region to obtain spatial information of a target in the entire field of view, whose quantity of sampling points is the same as that in uniform sampling method. This strategy achieves high-resolution reconstruction of the key area of the scene and ignores the useless information of surrounding area, which is able to improve the quality of images yielded by low-resolution sampling. Experiments were performed with a first-photon imaging system using the scanning galvo to achieve the sampling strategy at 50 × 50 pixel resolution. By comparing the experimental results by different sampling methods, it has been demonstrated that the proposed sampling method yielded images with the better reconstruction quality of the key area of the scenes with the same acquisition time. This method verified here could be useful for fast imaging fields, such as remote sensing.