Abstract: Time-correlated single-photon counting is a method for detecting weak light signals, offering photon-level detection accuracy and picosecond time resolution, and is widely used in lidar imaging. However, traditional imaging algorithms have certain limitations in edge information processing and photon utilization, which restrict the performance of single-photon lidar. To address this bottleneck, a dual-axis galvanometer scanning single-photon lidar imaging system was built based on the detection principle of single-photon lidar, and a Maximum A Posteriori estimation algorithm based on sparse regularization was proposed in this work. The results show that, compared with the traditional median filtering algorithm, the proposed imaging algorithm is more effective in noise reduction and detail preservation, with a 14.1% improvement in reflectivity image quality and a 51.9% improvement in depth image quality, while the image reconstruction time is reduced.

Key words: quantum optics, sparse regularization, time-correlated single-photon counting, 3D imaging; lidar