Back to Search Start Over

Retrieval of DTM under Complex Forest Stand Based on Spaceborne LiDAR Fusion Photon Correction

Authors :
Bin Li
Guangpeng Fan
Tianzhong Zhao
Zhuo Deng
Yonghui Yu
Source :
Remote Sensing, Vol 14, Iss 1, p 218 (2022)
Publication Year :
2022
Publisher :
MDPI AG, 2022.

Abstract

The new generation of satellite-borne laser radar Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) data has been successfully used for ground information acquisition. However, when dealing with complex terrain and dense vegetation cover, the accuracy of the extracted understory Digital Terrain Model (DTM) is limited. Therefore, this paper proposes a photon correction data processing method based on ICESat-2 to improve the DTM inversion accuracy in complex terrain and high forest coverage areas. The correction value is first extracted based on the ALOS PALSAR DEM reference data to correct the cross-track photon data of ICESat-2. The slope filter threshold is then selected from the reference data, and the extracted possible ground photons are slope filtered to obtain accurate ground photons. Finally, the impacts of cross-track photon and slope filtering on fine ground extraction from the ICESat-2 data are discussed. The results show that the proposed photon correction and slope filtering algorithms help to improve the extraction accuracy of forest DTM in complex terrain areas. Compared with the forest DTM extracted without the photon correction and slope filtering methods, the MAE (Mean Absolute Error) and RMSE (Root Mean Square Error) are reduced by 51.90~57.82% and 49.37~53.55%, respectively. To the best of our knowledge, this is the first study demonstrating that photon correction can improve the terrain inversion ability of ICESat-2, while providing a novel method for ground extraction based on ICESat-2 data. It provides a theoretical basis for the accurate inversion of canopy parameters for ICESat-2.

Details

Language :
English
ISSN :
20724292
Volume :
14
Issue :
1
Database :
Directory of Open Access Journals
Journal :
Remote Sensing
Publication Type :
Academic Journal
Accession number :
edsdoj.4890cc5ed98948efa9dfdb5429358397
Document Type :
article
Full Text :
https://doi.org/10.3390/rs14010218