A deformation field prediction method based on image quality evaluation and adaptive pre-training

By performing wavelet packet decomposition and quality assessment on the image, and combining supervised and unsupervised training of the U-Net model, the accuracy and robustness issues of the DIC model under extreme conditions are solved, achieving efficient and adaptive displacement and strain field prediction.

CN121811183BActive Publication Date: 2026-05-29HUNAN UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202610278943.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-03-09
Publication Date
2026-05-29
Estimated Expiration
2046-03-09

AI Technical Summary

Technical Problem

Traditional DIC models suffer from insufficient prediction accuracy and poor robustness under extreme conditions such as large deformation and speckle tearing. Furthermore, their reliance on high-quality labeled data leads to long training times, making it difficult to fully utilize unlabeled data and limiting the model's engineering applicability and scalability.

Method used

By performing three-level wavelet packet decomposition on the reference image and the deformation image, calculating the L2 norm and Pearson correlation coefficient to select subbands, reconstructing the image and calculating the quality evaluation index, and training the U-Net model using supervised and unsupervised training strategies, adaptive prediction of displacement and strain fields is achieved.

Benefits of technology

It significantly improves the engineering applicability and generalization ability of the prediction model in complex scenarios, enhances prediction accuracy and robustness, shortens training time, and strengthens the model's adaptability under extreme conditions.

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

Abstract

The application relates to a deformation field prediction method based on image quality evaluation and adaptive pre-training. Wavelet packet decomposition is performed on a reference image and a deformation image of a collected component to obtain coefficients of each subband of each layer of the images, and Pearson correlation coefficients of each subband are calculated based on all the coefficients corresponding to the subband. The subbands are screened based on L2 norms of coefficient vectors of the subbands and the Pearson correlation coefficients, and the screened coefficient vectors of the subbands are reconstructed to obtain reconstructed reference images and reconstructed deformation images. Quality evaluation indexes of each reconstructed sample pair are calculated. Training branches to which the corresponding reconstructed sample pairs belong are determined based on the quality evaluation indexes, a U-Net model is preliminarily trained by using a preset supervised training strategy based on the reconstructed sample pairs of different training branches, the U-Net model is unsupervisedly trained, and displacement fields and strain fields are predicted based on the trained U-Net model.
Need to check novelty before this filing date? Find Prior Art