大光斑激光雷达数据的森林冠层高度反演

Retrieval of forest canopy height based on large-footprint LiDAR data

针对传统的LM波形分解算法在GLAS大光斑波形数据处理中容易陷于局部最优解,限制了GLAS大光斑激光雷达数据在森林结构参数反演方面应用的问题,该文结合GLAS大光斑数据特征,引进优化后的EM算法对大光斑全波形数据进行分解,获取波形参数最优值。结合波形前缘长度和波形后缘长度,建立树高反演模型,并与LM分解算法建立的模型进行对比分析。研究结果表明,通过改进的EM算法对GLAS大光斑激光雷达数据进行处理,波形特征参数的获取更为精确,达到了较高的树高反演精度。

For the traditional LM waveform decomposition algorithm,it is easy to fall into the local optimal solution in GLAS large-spot data processing,which limits the application of GLAS large-footprint LiDAR data in the inversion of forest structural parameters.This paper combines the characteristics of GLAS large-footprint LiDAR data and introduces the modified EM algorithm to decompose the full-waveform data of large-footprint LiDAR data,and obtains the optimal value of waveform parameters.Based on the length of the leading edge and the length of the trailing edge,the inversion model of tree height was established and compared with the model established by LM decomposition algorithm.The results show that the improved EM algorithm can process the GLAS large-footprint LiDAR data,and obtain more precise waveform parameters,and achieve high inversion accuracy of tree height.