A Density-Based Adaptive Ground and Canopy Detecting Method for ICESat-2 Photon-Counting Data
Huan Xie, Dan Ye, Qi Xu, Yuan Sun, Peiqi Huang, Xiaohua Tong, Yalei Guo, Xiaoshuai Liu, Shijie Liu
Abstract
Ice, Cloud and land Elevation Satellite-2 (ICESat-2), the first photon-counting laser altimetry satellite, is the most advanced on-orbit altimetry system in the world. The data obtained by it contain a large number of background photons, limited by sensitive photon detection system. In this study, a density-based adaptive method (DBAM) for photons detection of ground and canopy method is proposed which is aimed at solving the problem of signal photons detection in vegetation areas. Firstly, the photon density is homogenized according to the noise photon rate, to reduce the effect of uneven background noise. And then the ground signal photons were extracted by the search ellipse adaptively changing direction and size along the slope direction to find the maximum density direction. The canopy signal photons were extracted again from the rest photons of the first step by using of a vertical elliptical search area. At last of the DBAM, the photons between the ground and the canopy are extracted as vegetation signal photons. The effectiveness of DBAM is evaluated both quantitatively and qualitatively. The results show that proposed method can effectively detect ground and canopy photons, with Kappa coefficient of 0.88 and 0.91 of two selected datasets, respectively. Compared to the results of ATL08, DBAM can better adapt to the slope, the extracted ground photons have better continuity, and can extract more accuracy vegetation photons.