A Method of Detecting Mis-Matching in Matching Between Scene Feature Points and Image Point Features

Abstract

The invention discloses a method for detecting mismatching in matching between a scene characteristic point and an image point characteristic. According to the method, a scene in which a dynamic object exists can be processed. The method comprises the following steps of: firstly, randomly sampling an image on the basis of priori to obtain sample data; then, dividing a sample point zone into inner points and outer points by the given hypothesis; evaluating the hypothesis by the amount and the distribution of the inner points until the hypothesis satisfies a given terminal condition; and finally, obtaining a model or data required for specific application by the obtained optimal hypothesis and an optimal sample. Even if a great quantity of mismatching or dynamic matching points are in the presence, a correct static matching point can be detected in real time. The method disclosed by the invention is obviously more superior to an existing mismatching detection method on two aspects of detection precision and operation efficiency.

Description

Technical field [0001] The present invention relates to a random sampling agreement (RANSAC) method, in particular to a method for detecting mismatches in the matching between scene feature points and image point features. Background technique [0002] Random sampling consensus (RANSAC) technology is a very useful tool for processing robust estimation problems in computer vision. In recent years, many researchers have continuously tried to improve the efficiency of random sampling consensus methods. [0003] The process of random sampling consensus method can be summarized as hypothesis formulation and hypothesis evaluation. In the hypothesis stage, a minimum sample set is obtained through random sampling, and a hypothetical model is fitted from the minimum sample set. In the hypothesis evaluation stage, all samples are used to evaluate the proposed hypothesis, and samples that meet the hypothesis are marked as inliers, and those that do not meet are marked as outliers. Repeat th...

AI Technical Summary

Problems solved by technology

However, the prior information of these methods fails to make full use of the correlation between images, so the obtained prior information is not sufficient. For example, the ProSAC method sorts the corresponding points (matching points) of adjacent images to Obtain prior information, because it is difficult to achieve accurate matching between corresponding points between adjacent images, and only select corresponding points with high matching degree, which may lead to sparse and unbalanced distribution of sampling points

[0006] 2. Existing Hypothesis Evaluation Methods

This method can terminate and obtain an optimal hypothesis in the specified time. However, in the application of the present invention, when the number of static points is small, it is also impossible to reliably select static matching points.

Images