Image data enhancement method and binocular stereo matching model training method and device

By generating random masks and disordered images based on target sample image pairs, data augmentation image pairs are constructed, which solves the problem of large differences between artificially defined noise and real noise, and achieves better training results and model fit.

CN116740494BActive Publication Date: 2026-06-23PING AN TECH (SHENZHEN) CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202310654138.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-02
Publication Date
2026-06-23
Estimated Expiration
2043-06-02

AI Technical Summary

Technical Problem

In existing image data augmentation methods, the artificially defined noise differs significantly from real noise, resulting in low fit with the training sample set and affecting the training effect of deep learning models.

Method used

By generating first and second random masks and out-of-order images, data augmentation images are generated based on target sample image pairs. The noise originates from the original image itself. Data augmentation sample image pairs are constructed to improve the fit between noise and the training sample set.

Benefits of technology

This improves the fit between noise and the training sample set, resulting in better training performance. It also expands the training sample set and enhances the robustness and generalization ability of the model.

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

Abstract

The application belongs to the technical field of medical treatment and can be used for sample data enhancement in the scene of human body, tissue organ modeling and the like in the medical field, and particularly relates to an image data enhancement method, a training method and device of a binocular stereo matching model. The method comprises the following steps: obtaining a target sample image pair, the target sample image pair comprising a first image and a second image; generating a first random mask and a second random mask; obtaining a first disordered image and a second disordered image; obtaining a first data enhancement image based on the first random mask, the second disordered image and the first image; obtaining a second data enhancement image based on the second random mask, the first disordered image and the second image; and combining the first data enhancement image and the second data enhancement image to obtain a data enhancement sample image pair. The above method and device improve the adaptability of the introduced noise to the training sample set, so that the expanded training sample has better training effect.
Need to check novelty before this filing date? Find Prior Art