Textured Digital Elevation Model Generation With Lidar Misregistration Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating textured digital elevation models from lidar and optical data face challenges in minimizing misregistration errors, leading to inaccuracies in the projection and alignment of lidar points and optical images across overlapping swaths.
Innovation Solution
A method that projects lidar point data into corresponding texel swaths with normalized coordinates, calculates enhanced lidar coordinates to minimize misregistration errors, and generates textured digital elevation models by creating enhanced output pixel coordinates based on these coordinates, thereby improving the alignment and accuracy of the model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional projection methods are used to map lidar points to optical images, then the processing speed is maintained, but misregistration errors increase leading to reduced accuracy
Solution Approach 1:
The patent applies preliminary action by pre-calculating projection matrices for each texel swath before the actual coordinate transformation. These projection matrices are computed in advance based on the swath geometry and sensor parameters, allowing the subsequent lidar point projection to use these pre-computed matrices rather than calculating projections in real-time. This preliminary preparation reduces misregistration errors while managing processing complexity through efficient pre-computation.
Solution Approach 2:
The patent replaces traditional mechanical or geometric projection methods with a matrix-based mathematical transformation system. Instead of using complex geometric calculations or iterative optimization during runtime, the system substitutes these with pre-computed projection matrices that perform coordinate transformations through efficient matrix multiplication. This substitution maintains processing speed while improving alignment accuracy through more precise mathematical transformations.
2Measurement precision
If enhanced coordinate calculations are performed to minimize misregistration errors, then alignment accuracy improves, but processing time increases
Solution Approach 1:
The patent applies segmentation by dividing the entire scene into multiple discrete texel swaths, each with its own projection matrix and coordinate system. Instead of performing enhanced coordinate calculations for all lidar points across the entire scene simultaneously, the system segments the processing into independent swath-based operations. This allows parallel processing of different swaths and reduces the computational burden per swath, thereby decreasing overall processing time while maintaining high coordinate accuracy through swath-specific optimization.
Solution Approach 2:
The patent applies partial action by performing enhanced coordinate calculations only for the specific region or swath being processed at each step, rather than recalculating coordinates for the entire scene repeatedly. The system processes lidar points in batches associated with individual texel swaths, applying enhanced coordinate transformations only where needed for each partial region. This partial processing approach reduces total processing time while maintaining accuracy through focused, swath-specific calculations.
Data Source
AI summary
For textured digital elevation model generation, a method projects swath point projection coordinates of each lidar point of texel swaths into a corresponding texel swath with normalized projection coordinates. Each texel swath includes lidar points, an optical image, and swath output projection point coordinates of the lidar points into the optical image. The method generates a projection matrix including lidar point data for each lidar point of each texel swath. The method calculates enhanced lidar coordinates for each lidar point as a function of minimizing misregistration errors. In addition, the method creates enhanced output pixel coordinates for each image pixel based on the enhanced lidar coordinates for the lidar points and the associated output pixel coordinates. The method generates a textured digital elevation model based on the enhanced output pixel coordinates for each image pixel, and the enhanced lidar coordinates.