InSAR interference network optimization method based on multi-factor coherence agent model

By constructing a multi-factor coherence surrogate model, and combining temporal baseline, spatial baseline and soil moisture factor, a highly coherent and highly redundant interferometric network was selected, which solved the problem of interferometric pair selection in massive SAR images and realized the efficient application of InSAR technology in large-scale, long-term and near-real-time monitoring.

CN122286409AActive Publication Date: 2026-06-26CHINA UNIV OF MINING & TECH (BEIJING)
4 Cites 0 Cited by

Patent Information

Application Number
CN202610270108.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-06
Publication Date
2026-06-26
Estimated Expiration
2046-03-06

AI Technical Summary

Technical Problem

Existing technologies struggle to adaptively and intelligently select high-quality interferometric pairs from massive SAR images, leading to the erroneous introduction of low-coherence interferograms and the erroneous removal of high-coherence interferograms. This affects the stability and accuracy of phase unwrapping and time-series inversion. Furthermore, the high computational and storage overhead limits the application potential of InSAR technology in large-scale, long-term, and near-real-time monitoring.

Method used

A multi-factor coherence surrogate model is constructed, which combines temporal baseline, spatial baseline and soil moisture influencing factors. The model is trained by randomly sampling a subset of interference pairs to predict coherence and select the optimal set of interference pairs. Graph theory is used to optimize the interference network, reduce computational overhead and improve coherence.

Benefits of technology

It effectively reduces the risk of erroneous introduction of low-coherence interferograms and erroneous removal of high-coherence interferograms, improves the stability and accuracy of untangling and time series inversion, reduces computational and storage overhead, and supports large-scale, long-time series and near-real-time monitoring needs.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses an InSAR (Interferometric Synthetic Aperture Radar) interference network optimization method based on a multi-factor coherence proxy model. The method comprises the following steps: constructing a multi-factor feature data set comprising an SAR image of a research area, baseline data and soil humidity data; constructing a candidate interference pair set, and randomly extracting from the candidate interference pair set as a sample interference pair subset; constructing a coherence agent prediction model, wherein the coherence agent prediction model performs model parameter training on the multi-factor feature data and the sample coherence truth value corresponding to the subset by using sample interference; and inputting the multi-factor feature data corresponding to each interference pair in the candidate interference pair set into the trained coherence proxy prediction model, outputting the predicted coherence of each interference pair, and optimizing and constructing an interference pair network. According to the method, the stability and precision of unwrapping and time sequence inversion are improved, meanwhile, the interference processing and coherence calculation difficulty under large-scale data is reduced, and the large-range, long-time-sequence and near-real-time monitoring requirements can be met.
Need to check novelty before this filing date? Find Prior Art