Forest Biophysical Parameter Estimation via SAR Temporal Decorrelation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating biophysical parameters of forests using Synthetic Aperture Radar (SAR) data are limited in efficiency and accuracy, particularly in capturing temporal changes and correlations.
Innovation Solution
A method utilizing temporally spaced SAR Single Look Complex (SLC) image acquisitions, processed through multitemporal SAR data processing to generate geocoded coherence images, followed by spatial averaging and application of a temporal decorrelation model to compute parameters like forest height and biomass, utilizing techniques such as speckle noise filtering and negative exponential models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional SAR data processing methods are used, then processing simplicity is maintained, but temporal decorrelation analysis accuracy deteriorates
Solution Approach 1:
The patent segments the SAR time series data into multiple coherence images representing different temporal baselines. Each coherence image is processed independently to extract temporal decorrelation characteristics, then these segmented results are integrated to estimate biophysical parameters. This segmentation enables precise temporal analysis while managing processing complexity through modular computation.
Solution Approach 2:
The patent transforms the temporal SAR data into a multi-dimensional coherence space by computing coherence between different time points. This dimensional transformation from temporal series to coherence magnitude-phase space enables extraction of temporal decorrelation information that is not apparent in the original temporal domain, improving measurement precision.
2Measurement precision
If single-time SAR images are used, then data processing time is reduced, but forest parameter estimation accuracy deteriorates
Solution Approach 1:
The patent performs preliminary processing of SAR images to generate coherence images before the actual parameter estimation. This preliminary action pre-computes the temporal relationships and decorrelation characteristics, so that the final parameter estimation uses pre-processed information, reducing the computational burden during the main estimation phase while maintaining high accuracy.
Solution Approach 2:
The patent utilizes continuous temporal SAR observations to compute coherence across multiple time points, maintaining continuous useful action in capturing forest changes. By continuously processing the temporal sequence and extracting decorrelation information at each step, the method achieves high estimation accuracy without requiring excessive processing time through efficient sequential computation.
3Measurement precision
If complex multitemporal processing procedures are applied, then temporal decorrelation analysis is improved, but processing efficiency deteriorates
Solution Approach 1:
The patent extracts only the essential temporal decorrelation information from the complex multitemporal SAR data by computing coherence magnitude and phase. Instead of processing all temporal dimensions in full complexity, it extracts the key decorrelation characteristics that directly relate to biophysical parameters, improving measurement precision while reducing unnecessary computational overhead.
Solution Approach 2:
The patent changes the processing parameters by working with coherence images derived from SAR data rather than raw SAR images. This parameter transformation simplifies the temporal decorrelation analysis by converting the problem into coherence domain, where decorrelation appears as magnitude changes, enabling more efficient processing while maintaining analytical precision.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A method, system, and computer programs for estimating biophysical parameters of forests are proposed. The method comprises: obtaining image acquisitions of a given area comprising forests from a data source, the image acquisitions comprising synthetic aperture radar, SAR, single look complex, SLC, images and being temporally spaced with each other; generating a stack of geocoded coherences images by submitting the obtained image acquisitions to a multitemporal SAR data processing procedure; selecting, for each geocoded coherence image in the generated stack, a number of sub-areas of the given area; grouping each of the selected sub-areas into a plurality of sub-area geocoded coherence stacks; spatially averaging each geocoded coherence image in the sub-area geocoded coherence stacks using a spatial averaging technique, obtaining a plurality of vectors as a result; and computing a parameter that depends on forest biophysical parameters by executing a temporal decorrelation model on the obtained plurality of vectors.