High efficiency first-photon imaging via adaptive non-uniform raster-scanning

With the emergence of single-photon detectors and time-correlated single-photon counting techniques, single-photon imaging has shown great potential in long-distance imaging, non-line-of sight imaging, underwater detection and fluorescence imaging, etc. In typical cases, single-photon imaging needs to accumulate at least one hundred photons per pixel to ensure the image quality. To overcome the drawback of time-consuming photon counting imaging, first-photon imaging was proposed to improve the imaging efficiency by recording only the first event of photon arrival at each spatial location in Lidar. However, because single-photon detection must be conducted under low detection probability, first-photon imaging cannot yet meet the requirements of high-speed applications. In this work, we propose an adaptive non-uniform raster-scanning method to further improve the imaging efficiency of the first-photon imaging. ’Adaptive’ means adaptively changing the optimal emitted pulses in different scenarios and adaptively changing the scanning intervals. The root mean square error of the reconstructed reflectivity image using the proposed method is only slightly higher than that of conventional first-photon imaging, while the detection time was shortened by nearly five times. The proposed method provides an alternative approach to improve the imaging efficiency of single-photon imaging detection, especially for applications involving sparse targets in large fields of view.