Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A method for image non-local stereo match

A stereo matching, non-local technology, applied in the field of image processing, which can solve the problems of sensitivity to noise and illumination changes, poor real-time performance, and insufficient consideration of weight support.

Active Publication Date: 2019-01-04
HOHAI UNIV CHANGZHOU
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Among them, the gradient information can highlight the edge of the image well, but it is very sensitive to noise and illumination changes
In the cost aggregation stage, most aggregation methods define a local window and solve the average cost value of pixels in the window. The

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
  • A method for image non-local stereo match
  • A method for image non-local stereo match
  • A method for image non-local stereo match

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0082] The flowchart of the specific embodiment of the present invention is as figure 1 As shown, an image non-local stereo matching method, the specific steps include:

[0083] 1) Combining color and gradient information as a similarity measure function to construct a cost calculation function. Specific steps include:

[0084] 11) Matching cost calculation is the basis of the entire stereo matching method. In fact, it is to measure the gray similarity under different disparities and calculate the initial matching cost value. The selection of the cost function will directly affect the accuracy of the final disparity map, so the selection has a higher The matching cost function of matching performance is the primary problem faced by stereo matching methods. The method of the present invention combines the two measures of color information and gradient information as the initial matching cost function. In order to avoid the mismatch between pixels with the same grayscale and ...

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 an image non-local stereo matching method, comprising the steps: color and gradient information are combined as a similarity measure function to construct a cost calculation function; the minimum spanning tree is constructed for the left and right images, and the cost function value is aggregated; Initial disparity map is obtained by WTA strategy, unstable points are obtained by left-right consistency detection, holes are filled, and then the edge of disparity map is detected; according to the obtained edge images, combined with the idea of pruning, the cost function values are further aggregated; finally, the final dense parallax map is obtained by parallax refinement. The method of the invention adopts RGB color information to avoid mismatch of pixel points of thesame gray level and different colors in the gray level image, and improves the matching precision of the weak texture area. The neighborhood of texture-free region can also provide corresponding weight support in the cost aggregation stage, which effectively solves the problem that texture-free region does not provide effective cost aggregation value.

Description

technical field [0001] The invention belongs to the technical field of image processing and relates to an image non-local stereo matching method. Background technique [0002] Stereo matching is an important hotspot and difficulty in computer vision. It uses the principle of stereo imaging to find the position of the corresponding point in the multi-eye image, restore the obtained two-dimensional image to the three-dimensional scene, and obtain accurate depth information if the matching is correct. This technology can be widely used in robotics, artificial intelligence, medicine, aerospace, industrial testing and other fields. In recent years, due to the diversification of application backgrounds, the requirements for matching accuracy have also been increasing. [0003] Scharstein gave a comprehensive overview of existing stereo matching algorithms, including four steps: matching cost calculation, cost aggregation, disparity calculation and disparity refinement. Stereo ma...

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): G06T7/55G06T7/33G06T7/90G06K9/62
CPCG06T7/337G06T7/55G06T7/90G06F18/24323
Inventor 霍冠英罗颖许金鑫
Owner HOHAI UNIV CHANGZHOU
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com