Mismatching Removal for Feature-Point Matching Based on Triangular Topology Probability Sampling Consensus

Zaixing He,Chentao Shen,Xinyue Zhao,Quanyou Wang,Huilong Jiang

doi:10.3390/rs14030706

Zaixing He, Chentao Shen + Show 3 more

Open Access

https://doi.org/10.3390/rs14030706

Copy DOI

Abstract

Feature-point matching between two images is a fundamental process in remote-sensing applications, such as image registration. However, mismatching is inevitable, and it needs to be removed. It is difficult for existing methods to remove a high ratio of mismatches. To address this issue, a robust method, called triangular topology probability sampling consensus (TSAC), is proposed, which combines the topology network and resampling methods. The proposed method constructs the triangular topology of the feature points of two images, quantifies the mismatching probability for each point pair, and then weights the probability into the random process of RANSAC by calculating the optimal homography matrix between the two images so that the mismatches can be detected and removed. Compared with the state-of-the-art methods, TSAC has superior performances in accuracy and robustness.

Highlights

Remote Sens. 2022, 14, 706. https://Feature-point matching is an important component of image processing, and it is widely used in remote sensing, including in target recognition, image registration, object tracking, pose estimation, etc
We propose a robust method for mismatching removal, namely, triangular topology probability sampling consensus (TSAC), which adapts to a high mismatching ratio
The method constructs a topological network of the feature points on the image and calculates the mismatching probability; We propose a new sampling method—probability sampling consensus—which weights the probability calculated above to the random process of the RANSAC so that the mismatches can be detected and removed

Summary

Introduction

Feature-point matching is an important component of image processing, and it is widely used in remote sensing, including in target recognition, image registration, object tracking, pose estimation, etc. With the applications of image feature-point matching becoming more diverse, its accuracy and robustness are of vital importance. There have been many types of research conducted on the optimization of feature-point matching recently. This kind of research can be divided into feature extraction and feature-point matching. Many of the feature points are similar in some applications, so mismatches that have a great influence on the accuracy for following the progress, such as image registration or pose estimation, do occur

Methods

Results

Discussion

Conclusion