Stereo Matching Method Based on Global Image Symmetric Correlation
A technology of stereo matching and correlation, applied in the field of stereo matching, can solve problems such as lack of symmetry, matching performance constraints, and failure to consider the correlation of the initial matching cost of the target image, etc., to solve the problem of low matching accuracy and wide application prospects Effect
Active Publication Date: 2020-08-14
TIANJIN UNIV
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, this kind of window aggregation stereo matching algorithm based on local correlation has the following disadvantages using the traditional local correlation initial matching cost calculation method: 1) The local window size needs to be specified, and the calculation of the initial matching cost does not take into account the pixels outside the local window 2) All pixels in the window contribute equally to the correlation calculation, without considering the brightness difference and spatial difference between pixels; 3) Only the initial matching cost correlation of the reference image is considered , does not take into account the initial matching cost correlation of the target image, and lacks symmetry
[0004] At present, in practical applications, in order to meet the real-time requirements of stereo matching, more stereo matching algorithms based on local window aggregation are usually used, but due to the limitations of the traditional initial matching cost, the matching performance is constrained
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0016] The stereo matching algorithm based on the symmetric correlation of global images of the present invention mainly consists of four parts: calculation of global correlation of reference images, calculation of global correlation of target images, calculation of initial matching cost of reference images, and generation of final disparity maps. The specific steps and principles are as follows:
[0017] 101: Use the left image as the reference image I R , the right image as the target image I T ;
[0018] 102: Reference image global correlation Z R calculation;
[0019] will refer to the image's I R The center pixel is denoted as (x R ,y R ), x R ,y R The horizontal and vertical coordinates of this pixel point respectively; (i R ,j R ) for I R any pixel in i R ,j R The abscissa and ordinate of this pixel respectively; Z R (x R ,y R ) for (x R ,y R ) is the global correlation value of the reference image at the central pixel point:
[0020]
[0021]
...
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
Login to View More Abstract
The present invention discloses a global image symmetry correlation-based stereo matching algorithm. The algorithm includes the following steps that: the correlation weight value WR (iR, jR) of a reference image is obtained, and the global correlation value ZR (xR, yR) of the reference image is calculated; the global correlation value ZT (xT, yT) of a target image is calculated; the initial matching cost eR (xR, yR) of the reference image is calculated; and the initial matching cost of the reference image is aggregated according to a window aggregation stereo matching algorithm, so that the final disparity map DR of the reference image is obtained.
Description
technical field [0001] The invention relates to stereo matching in the field of computer stereo vision, can be used for three-dimensional reconstruction of images, and provides guidance in medical images, media equipment and the like. Background technique [0002] Stereo matching technology is an important research direction in the field of computer vision, which aims to extract the depth information of the scene from the scene map obtained by the left and right cameras. Its application fields involve robot control, autonomous navigation and other scientific and technological fields. [0003] The traditional stereo matching algorithm usually adopts the algorithm based on window aggregation. By limiting the size of the local window, the initial matching cost in the window is aggregated to complete the final matching. The initial matching cost commonly used in the algorithm includes truncated brightness initial matching cost, truncated gradient initial matching cost, and corr...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/32G06T7/593
CPCG06T7/32G06T7/593
Inventor 朱程涛李锵滕建辅
Owner TIANJIN UNIV