对盖革模式APD激光雷达系统的距离像重构算法进行了研究, 设计了一种基于像素邻域核密度估计的重构算法。从系统原理出发, 结合探测概率模型研究了距离像重构算法的理论基础。根据系统特点提出了一种基于像素邻域核密度估计的改进算法, 并对其原理进行了分析。通过仿真数据对直方图算法和邻域核密度估计算法进行了验证, 以距离重构准确率曲线进行了定量评价对比, 并进一步将算法应用到真实盖革模式APD激光雷达数据中进行了距离像重构实验。实验结果表明, 在低帧数时, 基于像素邻域统计核密度估计的重构算法可有效提高距离像重构的效果。

The range image reconstruction algorithm of Geiger-mode APD laser radar system was studied, and a reconstruction algorithm based on pixel neighborhood kernel density estimation was designed. Starting from the system principle, the theoretical basis of the reconstruction algorithm of range image was studied with the detection probability model. According to the characteristics of the system, an improved algorithm based on pixel neighborhood kernel density estimation was proposed and its principle was analyzed. The histogram algorithm and the neighborhood kernel density estimation algorithm were verified by simulation data, and the range reconstruction accuracy rate curve was used for quantitative evaluation and comparison. The algorithm was further applied to real Geiger mode APD lidar data to reconstruct range image. The experimental results show that the reconstruction algorithm based on the statistical neighborhood kernel density estimation can effectively improve the reconstruction effect of the range image at low frame counts.