Unlock instant, AI-driven research and patent intelligence for your innovation.

Real-time matching method for binocular video images

A matching method, binocular video technology, applied in the field of image matching, can solve problems such as complex calculations, achieve the effect of expanding epipolar range, speeding up efficiency, and increasing the number of matching

Pending Publication Date: 2021-08-13
SOUTHEAST UNIV
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Among them, the SIFT algorithm has a good effect on anti-rotation and scaling, but its calculation is complex

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Real-time matching method for binocular video images
  • Real-time matching method for binocular video images
  • Real-time matching method for binocular video images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] The technical solution of the present invention will be further introduced below in combination with specific embodiments.

[0026] This specific embodiment discloses a real-time matching method of binocular video images, comprising the following steps:

[0027] Step 10) Use the FAST operator with a radius of 2 pixels to perform global feature point extraction on the binocular image to generate a feature point position matrix. The specific implementation steps are as follows:

[0028] The specific embodiment of the present invention adopts the ground images of 1920*1080 pixels in different orientations collected by two cameras at the same time as the original left and right images of the experiment, and the internal and external orientation elements of the two cameras are known. Distortion correction is performed according to formula (1):

[0029]

[0030] Among them, x', y' are the horizontal and vertical coordinates after distortion correction, x, y are the coord...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a real-time matching method for binocular video images. The method comprises the following steps of 1) performing global feature point extraction on a binocular image by using a FAST operator of which the radius is two pixels, and generating a feature point position matrix; 2) solving a left epipolar line on a left image and a corresponding epipolar line on a right image by using a coplanar condition method, and recording the positions of the feature points on the epipolar lines; and 3) matching the feature points on the epipolar lines of the left image on the right image one by one, performing search and correlation operation on the same-name epipolar lines and upper and lower adjacent epipolar lines of the right image during matching, and extracting the optimal matching points. According to the method, the feature point extraction speed is effectively increased by using the FAST with the radius of 2, and the number of correctly matched feature points is increased by adopting multi-epipolar line matching.

Description

technical field [0001] The invention relates to the field of image matching, in particular to an optimized FAST-based binocular image feature point matching method. Background technique [0002] Binocular image matching is a key technology in binocular vision, and it is also a technology that determines whether the 3D scene obtained by binocular vision is real. The more the number of correctly matched points with the same name is, the more realistic the obtained 3D scene information is. Binocular image matching follows the principle of epipolar line constraints, that is, the pixel on a certain epipolar line on the left image must be on the epipolar line with the same name on the right image. The search range of points with the same name is reduced, and the search efficiency is accelerated. [0003] The current feature matching algorithms mainly include SIFT, SURF, FAST, ORB and other algorithms. Among them, the SIFT algorithm has a good effect on anti-rotation and scaling...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00G06K9/46
CPCG06V20/653G06V10/40
Inventor 翁永玲彭震沙月进
Owner SOUTHEAST UNIV