GBSAR Displacement Monitoring via Amplitude Image Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ground-Based Synthetic Aperture Radar (GBSAR) interferometry faces limitations such as loss of coherence over vegetated areas, aliasing, atmospheric interference, and mono-dimensional displacement measurements, which restrict its ability to accurately and unambiguously estimate terrain and man-made feature displacements.
Innovation Solution
A method involving data pre-processing with incoherent temporal averaging, global image matching using amplitude images, and co-registration transformation to estimate displacements, allowing for non-ambiguous, 2D displacement measurements unaffected by atmospheric effects and leveraging corner reflectors for precise target tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If SAR interferometry is used to estimate displacements from GBSAR data, then high sensitivity to small displacements is achieved, but loss of coherence over time severely limits the method's applicability over vegetated and forested areas
Solution Approach 1:
The patent replaces the interferometric phase-based measurement system with an amplitude-based image matching system. Instead of relying on coherent phase information that degrades over time, the invention uses amplitude images and feature matching algorithms to measure displacements, thereby eliminating the coherence limitation while maintaining measurement precision
Solution Approach 2:
The patent changes the measurement parameter from interferometric phase to amplitude intensity values. By using amplitude-based features and image matching instead of phase differences, the system can operate over long time periods without suffering from coherence loss, enabling reliable displacement monitoring in vegetated areas
2Measurement precision
If SAR interferometry is used for displacement estimation, then sub-millimetre precision is achieved for good scatterers, but aliasing imposes a restrictive limit on unambiguous displacement estimation
Solution Approach 1:
The patent replaces the phase-wrapping interferometric measurement system with an amplitude-based image matching system. The feature matching approach tracks the position of identified features across multiple images, providing unambiguous displacement measurements without the aliasing constraints inherent in phase-based methods
Solution Approach 2:
The patent transitions from one-dimensional phase measurement to two-dimensional image space matching. By working with full amplitude images and matching features across the image plane, the system captures displacement information in multiple dimensions, eliminating the ambiguity inherent in single-phase measurements
3Measurement precision
If GBSAR interferometry is used to measure displacements, then the LOS component is obtained, but the mono-dimensional nature limits comprehensive 3D displacement characterization
Solution Approach 1:
The patent extends measurement from one dimension (LOS) to two dimensions by performing image matching in the full image plane. The system identifies features and tracks their positions across multiple images, enabling measurement of displacements in range and azimuth directions, thereby providing 2D displacement information
Solution Approach 2:
The patent creates a multi-functional measurement system that can extract multiple displacement components from the same dataset. By using amplitude-based feature matching on GBSAR images, the system can derive both range and azimuth displacement components, making the measurement system more versatile and adaptable to different monitoring needs
4Measurement precision
If standard GBSAR interferometry is used, then displacement measurements are obtained, but atmospheric heterogeneities degrade the quality of displacement estimation
Solution Approach 1:
The patent replaces the interferometric phase measurement system with an amplitude-based image matching system. Since atmospheric effects primarily impact phase information rather than amplitude information, the amplitude-based approach inherently eliminates atmospheric interference while maintaining displacement estimation quality
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 provides non-ambiguous, aliasing-free, and 2D displacement estimates, overcoming atmospheric interference and enhancing precision, with corner reflectors ensuring reliable measurements of target displacements over time.
Implementation Method 1
A Ground-Based Synthetic Aperture Radar (GBSAR) is an active microwave acquisition sensor that provides its own illumination and measures the signal reflected by targets
Implementation Method 2
The eventual displacements of the observed scene between two SAR acquisitions separated in time are derived by computing and analysing the interferometric phase, which is given, pixelwise, by the difference between the SAR phase acquired in the first campaign, minus the SAR phase of the second campaign
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for monitoring terrain and man-made feature displacements comprising pre-processing sets of complex SAR images acquired in a first and in a subsequent campaign by means of a GBSAR instrument to obtain an incoherent mean SAR image for each campaign; selecting, in each of the incoherent mean SAR images, pixels which are representative of targets respectively in each of said sets of SAR image data, the said targets being selected from artificial corner reflectors, if they are deployed in the area of interest, and natural reflectors in the area of interest; performing a global image matching over the selected pixels to obtain a set of pairs of global shifts expressed in pixels and selecting a plurality of quality-scored pairs of global shifts; estimating GBSAR repositioning effects by obtaining a mask comprising stable-area pixels corresponding to stable areas in the area of interest, computing a subset of global shifts falling in the mask, estimating a co-registration transformation on the grounds of the subset of global shifts, and producing a set of co-registration parameters modelling the GBSAR repositioning effects; estimating pairs of displacements for each campaign by subtracting the GBSAR repositioning effects modelled by the co-registration parameters from the set of pairs of global shifts, thereby obtaining a plurality of pairs of displacements shifts expressed in pixels which, converting the displacement shifts expressed in pixels into displacements expressed in a displacement unit, and obtaining a set of pairs of estimated displacements defined in an object space by geocoding involving an image-to-object transformation.