ICESat-2 Tree Height Mapping via Multi-Source Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for large-scale tree height mapping, particularly using LiDAR, face limitations in data resolution and coverage, especially with spaceborne LiDAR data, which hinders accurate and high-resolution forest canopy height assessment.
Innovation Solution
A method utilizing multi-source remote sensing data, including ICESat-2 LiDAR, airborne LiDAR, Sentinel-2 imagery, and ancillary data, involves preprocessing and calibration steps to establish a random forest regression model for generating 20-meter resolution forest tree height maps, incorporating photon-scale and spatial-scale calibrations to improve data accuracy and coverage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If spaceborne LiDAR is used for large-scale tree height mapping, then coverage area is improved, but measurement precision deteriorates
Solution Approach 1:
The patent combines spaceborne LiDAR data with airborne LiDAR data and optical remote sensing data to create a comprehensive tree height mapping system. This multi-source data integration allows the system to maintain large-scale coverage while improving measurement precision through data fusion and mutual validation.
Solution Approach 2:
The patent introduces airborne LiDAR data as an intermediary to bridge the precision gap of spaceborne LiDAR. By using airborne LiDAR (which has higher precision) to calibrate and validate spaceborne LiDAR measurements, the system achieves both large-scale coverage and acceptable precision.
2Productivity
If LiDAR data is used directly without calibration, then processing speed is improved, but measurement accuracy deteriorates
Solution Approach 1:
The patent performs preliminary calibration of spaceborne LiDAR data using airborne LiDAR data before final tree height mapping. This pre-calibration step establishes accurate reference relationships early in the process, ensuring measurement accuracy is maintained throughout subsequent processing operations.
Solution Approach 2:
The patent applies parameter calibration by transforming spaceborne LiDAR measurements through statistical relationships derived from airborne LiDAR data. This parameter adjustment process corrects systematic errors while maintaining the efficiency of spaceborne data processing.
3Measurement precision
If multi-source remote sensing data is integrated, then measurement accuracy is improved, but device complexity increases
Solution Approach 1:
The patent develops a unified processing framework that handles multiple data sources (spaceborne LiDAR, airborne LiDAR, optical imagery) through a single integrated algorithmic system. This multi-functional approach improves measurement accuracy through data fusion while avoiding the complexity of separate processing systems for each data type.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the accuracy and expressiveness of tree height data, achieves large-scale and dense coverage, and reduces errors by synergistic inversion of multi-source data, resulting in high-resolution forest canopy height mapping.
Implementation Method 1
Light Detection and Ranging (LiDAR) utilizing laser technology
Implementation Method 2
optical remote sensing imagery...gather spectral information
Data Source
AI summary
Based on ICESat-2 high-resolution data, the disclosure proposes a method for mapping tree height. The method includes acquiring ICESat-2 LiDAR data, airborne LiDAR data, Sentinel-2 imagery data, and ancillary data within the selected time and target area; preprocessing the Sentinel-2 imagery to calculate spectral feature parameters; performing photon-scale calibration on ICESat-2 LiDAR data, and combining with airborne LiDAR data for spatial-scale calibration to obtain tree height feature parameter; preprocessing the ancillary data and calculating terrain feature parameters and climate feature parameters; inputting the spectral feature parameters, terrain feature parameters, and climate feature parameters as independent variables, and the tree height feature parameter as the dependent variable into a random forest regression model to establish a forest tree height inversion model; and utilizing the forest tree height inversion model to generate forest tree height map at a resolution of 20 meters within the target area.