Abstract：Full waveform hyperspectral LiDAR can simultaneously obtain three-dimensional spatial information and spectral information of targets. As a result of it, target attributes in the vertical direction can be accurately assessed. However, spectral reflectance extraction from full waveform hyperspectral LiDAR data, especially for multi-echo full waveforms, is still challenging due to the indeterminate backscattering cross-sectional area and unseparated reflectance. This challenge hampers the extraction accuracy of multi-echo reflectance and then causes underestimation consequently. Here, we proposed a new method to extract reflectance of multi-echo full waveforms, namely multi-echo reflectance correction using neighbourhood single-echo reflectance (MCNS), and verified its validation by using the coefficient of determination ( R 2 ) and the root mean square error (RMSE) based on simulated and measured data. Results indicated that R 2 and RMSE between the corrected simulated multi-echo reflectance and the spectrometer reflectance were more than 0.95 and less than 0.042 respectively, while that between the corrected measured multi-echo reflectance and the spectrometer reflectance were above 0.56 and below 0.15 respectively. These results demonstrated the feasibility of the proposed method in extracting multi-echo reflectance. This study lays a foundation for subsequent forest attribute assessments more accurately, making it possible to characterise multiple target features vertically using high-resolution spectral reflectance.
KeyWord：Full waveform hyperspectral LiDAR; Backscattering cross-sectional area; Multi-echo reflectance; LESS;

相关文献：
1.Knowledge evolution learning: A cost-free weakly supervised semantic segmentation framework for high-resolution land cover classification
2.geeSEBAL-MODIS: Continental-scale evapotranspiration based on the surface energy balance for South America
3.Suaeda salsa spectral index for Suaeda salsa mapping and fractional cover estimation in intertidal wetlands
4.Better localized predictions with Out-of-Scope information and Explainable AI: One-Shot SAR backscatter nowcast framework with data from neighboring region
5.Blurry dense object extraction based on buffer parsing network for high-resolution satellite remote sensing imagery
6.Rapid survey method for large-scale outdoor surveillance cameras using binary space partitioning
7.Edge aware depth inference for large-scale aerial building multi-view stereo
8.An efficient point cloud place recognition approach based on transformer in dynamic environment
9.Unsupervised domain adaptation for SAR target classification based on domain- and class-level alignment: From simulated to real data