Research on triggering properties enhancement of polarization detection geiger-mode APD LIDAR

Abstract

Gm-APD (Geiger-mode avalanche photodiode) is widely used in the LIDAR (Light Detection and Ranging) for its excellent photon detection capability. However, the low SNR (Signal-to-Noise Ratio) limits the laser echo's triggering probability and enhances the false alarm probability in the daytime, which severely affects the target's detection results, moreover, it increases the difficulty of the signal reconstruction algorithm. So it's an important path to relieve the algorithm's burden and get a better detection results by fundamentally increasing the echo's SNR. Nevertheless, there is a lack of researches discussing such issue, especially for array imaging detection. In this paper, it was proposed that by using the polarization detection could help to enhance the echo's SNR and get good detection results in the daytime. By combining polarization bi-directional reflection distribution function (pBRDF), Mueller matrix, LIDAR equation, and Poisson models, the polarization Gm-APD LIDAR detection model was first established to describe the received Stokes vector as well as the triggering and false alarm probabilities, which could aid in the daytime polarization detection including polarization analysis, depolarization contrast analysis, detection probabilities analysis, and the feasibility of the models was supported by the simulation and experimental results. Based on the established model, the triggering properties of the Gm-APD LIDAR with the polarization system were discussed, and the theoretical analysis showed that the optimal detection angle was 135°, at which the polarization detection could effectively increase the triggering probability and lower the false alarm probability. A series of outdoor experimental verifications were carried out, and the experimental results showed that the proposed Gm-APD polarization detection model and the theoretical analysis were absolutely correct. Besides, compared with the non-polarization detection, the intensity image of optimal polarization detection could completely obtain the targets’ contour, moreover, the triggering probabilities of two targets increased by 4.9% and 7.1%, respectively, and the false alarm probabilities lowered by 26.3% and 21.4%. It made great improvement in the Gm-APD detection research. This work offers the theoretical and experimental supports for enhancing Gm-APD LIDAR's detection results in the daytime by using polarization detection, furthermore, it can lay some theoretical foundations for target recognition of Gm-APD polarization detection.