SAR Geolocation Using Orthogonal Acquisitions and Tropospheric Delay Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for absolute 3D urban reconstruction using spaceborne Synthetic Aperture Radar (SAR) imagery require a large number of images, are costly, time-consuming, and rely on ground control points and external data for accurate signal path delay estimation, which can be inaccurate and labor-intensive.
Innovation Solution
A method utilizing two long-aperture SAR acquisitions with different geometries for automatic 3D geolocation and simultaneous estimation of tropospheric propagation delays, eliminating the need for ground control points and external data, by mapping point targets between images and solving equations to derive precise absolute heights and tropospheric delays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large set of SAR images is used for 3D urban reconstruction, then the accuracy of the reconstruction is improved, but the acquisition time and costs increase significantly
Solution Approach 1:
The invention changes the geometric parameters of the SAR acquisitions by using two images with significantly different viewing angles (orthogonal geometries). This parameter change allows the system to achieve sufficient geometric diversity for accurate 3D reconstruction without requiring a large temporal baseline, thus reducing acquisition time from approximately one year to a much shorter period while maintaining reconstruction accuracy.
2Measurement precision
If ground control points are used for accurate 3D geolocation, then the positioning accuracy is improved, but the complexity and cost of the procedure increase due to dedicated ground surveys
Solution Approach 1:
The system performs self-calibration by automatically determining its own geometric parameters and tropospheric delays from the SAR image data itself, without requiring external ground control points. The method uses the redundancy in the two orthogonal observations to solve for unknown parameters including baseline vector, focal length, and atmospheric delays, making the system self-sufficient and eliminating the need for complex ground surveys.
3Reliability
If external data and models are used to estimate tropospheric delays, then the signal path delay correction is improved, but the accuracy is reduced due to model projection errors to radar line of sight
Solution Approach 1:
The system uses feedback from the observed point target positions in the two orthogonal SAR images to iteratively refine the tropospheric delay estimates. By comparing the observed geometric relationships with those predicted by the current delay estimates, the system adjusts the delay parameters to minimize discrepancies, thereby improving positioning accuracy without relying on external models.
4Extent of automation
If SAR radargrammetry is used for automatic point target identification, then the automation level is improved, but the accuracy of identifying point targets in different geometries is reduced
Solution Approach 1:
The invention exploits the asymmetric geometric relationship between two orthogonal SAR acquisitions to uniquely identify point targets. By using the different viewing angles, the system creates distinctive imaging geometries for each target, allowing automatic identification and matching of corresponding points across the two images without ambiguity, thereby maintaining high accuracy in automated processing.
Data Source
Figure 1
Figure 2~3
Figure 4a~5b
AI summary
The invention describes a method for the automatic three-dimensional geolocation of SAR targets and simultaneous estimation of tropospheric propagation delays using two long-aperture SAR images, comprising - two SAR acquisitions with long aperture acquired with different geometry, - point target detection independently in both images, and - point target defocus analysis, - wherein for each point target a relationship between estimated height and the tropospheric delay is established.