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

Fast stereo matching method based on color partitioning and self-adaptive window

An adaptive window and color segmentation technology, which is applied in the field of computer vision, can solve the problems of high computational cost, inability to reflect the efficiency of local algorithms, and complex calculations, and achieve the effect of quickly obtaining high-precision dense disparity maps

Inactive Publication Date: 2011-07-20
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF4 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These methods propose an adaptive weight method based on the color and geometric relationship between each pixel in a given window and the pixel to be matched, which can effectively construct the matching cost, greatly reduce the ambiguity of matching, and obtain the disparity The results can be compared with the global optimization results, but this method is computationally complex and expensive, and cannot reflect the advantages of high efficiency of local algorithms

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
  • Fast stereo matching method based on color partitioning and self-adaptive window
  • Fast stereo matching method based on color partitioning and self-adaptive window
  • Fast stereo matching method based on color partitioning and self-adaptive window

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] Below in conjunction with accompanying drawing and example the patent of the present invention is further described.

[0025] like figure 1 As shown, a fast stereo matching method based on color segmentation and adaptive window includes the following steps:

[0026] In the first step, under a certain disparity assumption d, calculate the initial matching cost δ(p, q) of the corresponding matching pixel:

[0027] δ(p,q)=|R p -R q |+|G p -G q |+|B p -B q | (1)

[0028] In formula (1), p∈I r and q∈I t is a pair of matching points under the disparity assumption d.

[0029] In the second step, use the Mean Shift algorithm to segment the reference image and the target image, and respectively construct the matching cost C of the corresponding matching points p and q based on the color segmentation of the reference image under the disparity d sr and the matching cost C based on the color segmentation of the target image st :

[0030] C ...

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 fast stereo matching method based on color partitioning and self-adaptive window, comprising the following steps of: (1) constructing a matching cost based on color partitioning; (2) constructing a matching cost based on self-adaptive window; (3) constructing a combined matching cost based on the color partitioning and the self-adaptive window; and (4) optimizing the combined matching cost by adopting a local optimization method in order to obtain a dense disparity map. The method of the invention not only can improve the matching precision of parallax discontinuous areas and low veins areas, but also obtains the parallax which is comparable with that of current dominant algorithms. In addition, the processing time of the disclosed method is enhanced by about 19 to 35 times compared with the current excellent local methods, and both the matching speed and the matching precision are taken into consideration.

Description

technical field [0001] The invention relates to a stereo matching method in a stereo vision system, belongs to the field of computer vision, and is mainly applied to quickly obtain high-precision dense parallax information, thereby providing a basis for three-dimensional information restoration based on stereo vision. Background technique [0002] Stereo matching is an important task in the field of computer vision. It obtains the disparity map through binocular or multi-eye image matching, so as to perceive the three-dimensional depth information in the scene. Many scholars at home and abroad have conducted in-depth research on this field. Among them, the local stereo matching algorithm has high efficiency and is easy to implement, but it always has the problem of how to adaptively determine the size and shape of the support window, which makes it difficult to achieve high matching accuracy. Aiming at this problem, many scholars have proposed many representative local matc...

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): G06K9/62
Inventor 倪炜基徐贵力李文跃王彪倪立学
Owner NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
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