A Method of Arc Matching Based on Feature Descriptor

Abstract

The present invention proposes a feature descriptor-based arc matching method for images with arc features under different transformations. First, use the ELSD method to extract arcs from the original image; then represent the arcs as arc strips, generate arc support areas, construct arc descriptors (CBD) in the support areas, and obtain the band description matrix (BDM); finally Geometric attributes and minimum descriptor distance constraints are used to generate candidate matching pairs, and an adjacency matrix is ​​used to establish a relationship graph between two sets of candidate matching pairs to obtain the final arc matching result. A large amount of data analysis results prove that the arc matching method based on feature descriptors proposed by the present invention has a larger number of correct matches and a higher matching accuracy rate under translation, rotation, scaling transformation and illumination changes. It is of great significance to the development of image processing, machine vision, autonomous navigation and other fields.

Description

technical field [0001] The invention relates to the technical field of image matching, in particular to an arc matching method based on feature descriptors. Background technique [0002] Images are generally composed of basic elements such as points, lines, and surfaces. Image matching has important applications in 3D reconstruction, target recognition, target tracking, and visual navigation. Generally, image matching based on feature descriptors involves three steps: feature Extract, describe and match. [0003] In the prior art, there have been many studies on point matching, straight line matching, curve matching and area matching, but there is little research on arc matching, especially on arc matching based on feature descriptors. So far, the arc extraction algorithms involved include Hough transform (HT), EDCircles, etc., but the EDCircles algorithm is based on the edge detection with the EDPF (Edge DrawingParameter Free) algorithm, and then uses the EDLines algorithm...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, there have been many studies on point matching, straight line matching, curve matching and area matching, but there is little research on arc matching, especially on arc matching based on feature descriptors.

So far, the arc extraction algorithms involved include Hough transform (HT), EDCircles, etc., but the EDCircles algorithm is based on the edge detection with the EDPF (Edge DrawingParameter Free) algorithm, and then uses the EDLines algorithm to convert the edge into a straight line, and finally Combining two heuristic algorithms to convert straight lines into arcs, the extraction efficiency and accuracy are difficult to guarantee

[0004] At present, arc matching mainly has the following difficulties: 1) Unlike point matching, it is easier to select a fixed-size area to calculate the descriptor. Arcs of different lengths have different sizes of support areas, and there is no suitable area normalization strategy to calculate the area. Descriptor; 2) The arc description is more difficult than the straight line description, and the different positions of the arc have different pixel distribution rules, which exacerbates this difficulty; 3) Unlike the epipolar line constraint of point matching, which can provide strong global constraints, The epipolar line can hardly directly provide effective position constraints for arc matching, which increases the search space for arc matching and also increases the possibility of mismatching.

[0005] At home and abroad, there is no relevant report on arc matching based on feature descriptors

Images