Image feature point matching method and device, and storage medium

By using network graphs and maximum flow/minimum cut algorithms in image feature point matching, the correct matching point pairs are determined based on disparity information, which solves the problems of threshold dependence and insufficient generalization ability in existing methods, and achieves higher matching accuracy and wider applicability.

CN115482403BActive Publication Date: 2026-07-17CHINA COAL RES INST
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
CHINA COAL RES INST
Filing Date
2022-09-01
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing feature point matching methods rely on threshold settings and have insufficient generalization ability, resulting in inaccurate matching results, especially in scenarios where they have not been trained.

Method used

By identifying multiple candidate matching point pairs of feature points in the image, a network graph is mapped, and an energy function is established based on image disparity information. The minimum cut of the network graph is solved using the maximum flow/minimum cut algorithm to determine the correct and incorrect matching point pairs, thus avoiding threshold setting.

Benefits of technology

It improves the accuracy of feature point matching and scene generalization ability, reduces the dependence on threshold setting, and improves the matching effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115482403B_ABST
    Figure CN115482403B_ABST
Patent Text Reader

Abstract

The present disclosure provides a feature point matching method and device for images and a storage medium. The method comprises determining a plurality of candidate matching point pairs of feature points in two images, mapping the plurality of candidate matching point pairs into a network graph, establishing a first energy function representing the energy required for marking the plurality of candidate matching points in the network graph as correct matching point pairs or incorrect matching point pairs based on image disparity information, solving the minimum cut of the network graph as the optimal value of the first energy function by using a maximum flow / minimum cut algorithm, and determining the correct matching point pairs and the incorrect matching point pairs according to the minimum cut. The correct matching point pairs in the images can be determined by using the graph cut principle. This process does not need to rely on the setting of a threshold, thereby improving the accuracy of feature point matching. In addition, the present scheme does not need to perform model training and has strong scene generalization ability. Thus, the effect of feature point matching is improved.
Need to check novelty before this filing date? Find Prior Art