This study proposes a complete data-processing framework for Geiger-mode avalanche photodiode (GM-APD) light detection and ranging (LiDAR) echo signals. It investigates the feasibility of classifying target and background noise using machine learning. Four feature processing schemes were first compared, among which the PNT strategy (Principal Component Analysis without tail features) was identified as the most effective and adopted for subsequent analysis. Based on this framework, nine models derived from six baseline algorithms—Decision Trees (DTs), Support Vector Machines (SVMs), Backpropagation Neural Networks (NN-BPs), Linear Discriminant Analysis (LDA), Logistic Regression (LR), and k-Nearest Neighbors (KNN)—were systematically assessed under Monte Carlo simulations with varying echo signal-to-noise ratio (ESNR) and statistical frame number (SFN) conditions. Model performance was evaluated using eight metrics: accuracy, precision, recall, FPR, FNR, F1-score, Kappa coefficient, and relative change percentage (RCP). Monte Carlo simulations were employed to generate datasets, and Principal Component Analysis (PCA) was applied for feature extraction in the machine learning training process. The results show that LDA achieves the shortest training time (0.38 s at SFN = 20,000), DT maintains stable accuracy (0.7171–0.8247) across different SFNs, and NN-BP models perform optimally under low-SNR conditions. Specifically, NN-BP-3 achieves the highest test accuracy of 0.9213 at SFN = 20,000, while NN-BP-2 records the highest training accuracy of 0.9137. Regarding stability, NN-BP-3 exhibits the smallest RCP value (0.0111), whereas SVM-3 yields the largest (0.1937) at the same frame count. In conclusion, NN-BP-based models demonstrate clear advantages in classifying sky-background noise. Building on this, we design a ResNet based on NN-BP, which achieves further accuracy gains over the best baseline at 400, 2000, and 20,000 frames—12.5% (400), 9.16% (2000), and 2.79% (20,000)—clearly demonstrating the advantage of NN-BP for GM-APD LiDAR signal classification. This research thus establishes a novel framework for GM-APD LiDAR signal classification, provides the first systematic comparison of multiple machine learning models, and highlights the trade-off between accuracy and computational efficiency. The findings confirm the feasibility of applying machine learning to GM-APD data and offer practical guidance for balancing detection performance with real-time requirements in field applications.

Photon-efficient LiDAR provides an active remote sensing pathway for long-range, low-light three-dimensional observation, but photon-limited reconstruction remains difficult when background events dominate sparse returns. In dynamic scenes, line-of-sight motion further disperses weak-target photons across range-time bins and broadens fixed-bin accumulation. Existing methods mainly exploit histogram statistics, spatial priors, or learned representations, whereas explicit range-time migration constraints for interframe photon alignment remain insufficiently explored. We present a range-time trajectory-guided photon accumulation method for a 1064 nm InGaAs/InP $64\times 64$ GM-APD array. The method combines range-dependent background correction, multi-scale photon statistics, bounded trajectory parameter optimization using PSO, trajectory-aligned accumulation, and confidence-weighted spatial-range fusion. At -12.34 dB in simulation, SSIM, RMSE (range bins), PSNR, IoU, and FPR were 0.61, 223.22, 6.36 dB, 0.74, and 0.10%, respectively. Across 19 experimentally generated low-SBR evaluation cases derived from measured GM-APD LiDAR sequences, the corresponding mean values were $0.541\pm 0.165$, $332.93\pm 79.60$, $8.60\pm 2.49$ dB, $0.702\pm 0.061$, and $0.33\pm 0.08$%. Under the tested conditions, range-time guidance preserved spatial structure and pixel-wise range information while suppressing background responses, supporting photon-limited reconstruction in active remote sensing.