Balanced semi-dense stereo matching method

Abstract

The present invention belongs to the computer vision technical field and discloses a balanced semi-dense stereo matching method. The invention aims to improve matching efficiency while achieving higher image matching accuracy. The method includes the following steps that: a, two images to be matched are segmented; b, pre-feature extraction initialization setting is performed; c, points of interest are extracted from image blocks: in the extraction process, a feature extraction threshold value is adaptively adjusted, merged image blocks are selected according to needs, and points of interest evenly distributed in the images are obtained; e, interest of points in the two images are matched; e, a Delaunay triangle is constructed according to the matched interest of points; and d, the parallax of all the other non interest of points is estimated according to the Delaunay triangle, and thus image stereo matching is completed. The method of the invention is particularly suitable for image matching in a mobile platform or application fields having high real-time requirements.

Description

technical field [0001] The invention relates to computer vision technology, in particular to a balanced semi-dense stereo matching method. Background technique [0002] Image stereo matching is an important branch of computer vision, photogrammetry, and computer graphics, and it is of great value in many applications. Image matching can be divided into sparse matching and dense matching. Sparse matching generally extracts the feature points with strong texture on the image, and then calculates the matching cost through the feature descriptor to obtain the optimal matching. Due to the sparsity of feature points, sparse matching cannot provide a sufficient number of feature points and 3D points in many applications, so only limited 3D world information can be obtained. Dense matching is to match each pixel of the image, so dense 3D world information can be obtained. Algorithms for dense matching can be divided into global methods and local methods. In recent years, with th...

AI Technical Summary

Problems solved by technology

However, due to its inherent computational limitations, dense matching cannot achieve higher processing frame rates (eg, VGA image efficiency of 30fps or higher under equivalent computing conditions)

[0003] Therefore, in the traditional technology, on the one hand, sparse matching has high processing efficiency, but cannot provide a sufficient amount of matching point information; on the other hand, dense matching can obtain dense matching point information, but cannot obtain high processing efficiency, and cannot adapt to Higher and higher industry requirements

Images