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

Parallax correction method based on super pixel segmentation

A super-pixel segmentation and parallax correction technology, applied in the field of parallax correction, can solve the problems of foreground expansion, high complexity, unsuitable for real-time requirements, etc., and achieve the effect of high speed and high precision

Active Publication Date: 2017-05-10
成都快眼科技有限公司
View PDF5 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the local stereo matching algorithm, although the algorithm complexity of the fixed window stereo matching algorithm is low, there are problems of foreground expansion and poor parallax effect in the depth discontinuity area, while the stereo matching algorithm of adaptive window and adaptive weight greatly reduces the Mismatching in depth discontinuous areas, but the complexity is high, not suitable for occasions with high real-time requirements

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
  • Parallax correction method based on super pixel segmentation
  • Parallax correction method based on super pixel segmentation
  • Parallax correction method based on super pixel segmentation

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment 1

[0026] Such as figure 1 As shown, a disparity correction method based on superpixel segmentation is divided into two stages: initial disparity calculation and disparity correction.

[0027] The initial disparity calculation consists of the following four steps:

[0028] S1. For the left and right images captured by the corrected binocular camera, perform superpixel block segmentation on the left and right images respectively according to the pixel brightness values;

[0029] S2. Calculate the gradient of the input image after the superpixel block segmentation, and use the gradient and brightness as the calculation

[0030]

[0031]

[0032] C(x,y,d)=(1-ω)*C SAD (x,y,d)+ω*C GRAD(x,y,d) ...................................(3)

[0033] Parallax matching cost:

[0034] Among them, N(x, y) is the superpixel block area to which the matching point (x, y) belongs. represents the horizontal component of the image gradient, Represents the vertical component of the image gr...

specific Embodiment 2

[0045] On the basis of the specific embodiment 1, in said S5, calculate the normalized histogram of the current superpixel block and its adjacent superpixel blocks, and compare the histogram distance between the current superpixel block and the adjacent superpixel blocks as A measure of brightness similarity.

[0046] Specific implementation 3

[0047] On the basis of specific embodiment 1 or 2, the reliability threshold is 0.9.

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 provides a parallax correction method based on super pixel segmentation. In view of a left image and a right image photographed by a binocular camera after correction, super pixel block segmentation is respectively carried out on the left image and the right image according to pixel brightness values; in view of a non-texture region mismatching problem, instable super pixel blocks are marked for super pixel blocks in a flat area; and according to the fact that the parallaxes are closer in regions in which the adjacent super pixel blocks are more similar, the parallax of an instable super pixel block is corrected. In view of the foreground inflation problem of a local stereo algorithm, based on super pixel segmentation, the mismatching regions are corrected according to the fact that the parallaxes are closer in regions in which the adjacent super pixel blocks are more similar, the parallax map generation speed is quick, and the precision is high.

Description

technical field [0001] The invention relates to a parallax correction method, in particular to a parallax correction method suitable for superpixel segmentation. Background technique [0002] Most current stereo matching algorithms include four steps: (1) cost calculation, (2) cost aggregation, (3) disparity calculation and optimization, and (4) disparity refinement. Generally speaking, stereo matching algorithms can be divided into two categories: local stereo matching algorithms and global stereo matching algorithms. Local stereo matching algorithms use color or grayscale information within a window to determine the disparity value for each point. The global stereo matching algorithm is based on smoothness assumption, and uses energy minimization technique to determine the disparity value of all points simultaneously. [0003] The global stereo matching algorithm solves the minimum value of the overall energy by using two constraints, the image smoothness item and the da...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/136G06T7/11
CPCG06T2207/10012G06T2207/20021G06T2207/30168
Inventor 李宏亮孙文龙王久圣廖伟军罗旺
Owner 成都快眼科技有限公司