SAR Georeferencing of Digital Elevation Models Without Ground Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for geometric calibration of digital elevation models, such as those derived from photogrammetry and radar interferometry, suffer from inaccuracies due to uncertainties in sensor orientation and require ground control points, limiting absolute accuracy.

Innovation Solution

A method involving the use of synthetic aperture radar images for geometric calibration, including the selection of areas of interest, simulation of radar images, estimation of geometric offsets, and stereoscopic triangulation to correct and realign digital elevation models without relying on ground control points.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If ground control points are used for geometric calibration, then measurement precision is improved, but device complexity and operational difficulty increase

Engineering Contradiction:
Improveabsolute accuracyVSAvoidcomplexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The invention extracts and removes the requirement for ground control points from the calibration process. By using simulated radar images generated from the digital elevation model itself, the method eliminates the need for external reference data, thereby maintaining measurement precision while reducing operational complexity and eliminating the need for field surveys.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The invention creates simulated radar images that are copies or representations of what the radar sensor would observe from the digital elevation model. These simulated images serve as reference data for calibration, replacing the need for actual ground control points while maintaining the calibration functionality.

Inventive Principle:
Principle #26Copying

2Measurement precision

If ground control points are used for geometric calibration, then measurement precision is improved, but ease of operation deteriorates

Engineering Contradiction:
Improveabsolute accuracyVSAvoidease of operation
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The digital elevation model performs self-calibration by generating its own simulated radar images from its elevation data. This self-service approach eliminates the need for external ground control points and field surveys, making the calibration process entirely automated and significantly easier to operate while maintaining high measurement precision.

Inventive Principle:
Principle #25Self-service

3Productivity

If sensor orientation uncertainty is present, then productivity is maintained, but measurement precision deteriorates

Engineering Contradiction:
ImproveproductivityVSAvoidabsolute accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The invention implements a feedback mechanism where simulated radar images are generated from the digital elevation model and compared with actual radar observations. The discrepancies between simulated and observed data provide feedback for iteratively optimizing sensor orientation parameters, thereby correcting orientation uncertainties and improving measurement precision without reducing productivity.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4139707B1Method for georeferencing of a digital elevation model
Publication Date: 2025.11.26 AIRBUS DEFENCE & SPACE SAS
  • EP4139707B1 patent drawingFigure 1~2
  • EP4139707B1 patent drawingFigure 3~4
  • EP4139707B1 patent drawingFigure 5~6

AI summary

The invention relates to a method (100) for referencing a digital elevation model (10), involving a step of obtaining (110) two SAR images (12, 14) having a common area (15) that their area of overlap (16, 17) shares with the digital model (10); the method (100) involving, for the SAR images (12, 14), steps of: selecting (120) an AOI (18) in the common area (15); calculating (130) a simulated image (22) in the AOI (18); estimating (140) an offset (di, dj) between the simulated image (22) and the SAR image (12); the method (100) involving steps of selecting a reference point (26) in the AOI (18); projecting (160) the reference point (26) into the SAR images (12, 14) to obtain one connection point per SAR image (12, 14); correcting (170) the connection points by the offsets (di, dj); calculating (180) the readjusted reference point (26); referencing (190) the digital model (10).