Rapid stereo matching method based on ZSAD-Census

A stereo matching and fast technology, applied in the field of computer vision, can solve the problems of high power consumption and high area cost, and achieve the effect of increasing matching accuracy, strong anti-interference and high stability

Pending Publication Date: 2021-10-29
南京物间科技有限公司
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above problems, the present invention proposes a fast stereo matching method based on ZSAD-Census, which solves the problems of large power consumption and high area cost in chip implementation in the existing stereo matching algorithm, and improves the stereo matching algorithm at the same time. accuracy and anti-interference

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
  • Rapid stereo matching method based on ZSAD-Census
  • Rapid stereo matching method based on ZSAD-Census
  • Rapid stereo matching method based on ZSAD-Census

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0072] This application proposes a fast stereo matching method based on ZSAD-Census, including the following steps:

[0073] Such as figure 1 As shown, S1. Before the start of the stereo matching of each frame image, set the Census and grayscale of the stereo matching cost calculation to use the same neighborhood range, the neighborhood window height is 15, the width is 15, and the Census information and ZSAD information are both set Use a window of 5×5, configure the weights in the matching cost calculation formula as λ1=0.65 and λ2=0.35, set the search range of stereo matching as a window of 256×9, and set hash table 1 HTable and hash table 2 The maximum search times of HTable2 are 8 and 32 respectively.

[0074] Such as Figure 1-2 As shown, S2. Input the target image and reference image and traverse each pixel in the image, using N h ×N v Neighborhood gray information, calculate ZSAD information, using M h × M v Neighborhood gray information, calculate Census informa...

Embodiment 2

[0125] The present application also proposes a fast stereo matching method based on ZSAD-Census, including the following steps:

[0126] Such as Figure 1-2 As shown, S1. Before the stereo matching of each frame of image starts, according to the statistical information of the existing frame, configure ZSAD information and Census information to use a 5×5 window to generate, and configure the window of Census data when calculating the matching cost 15×15, configure the window of grayscale data used in the matching cost calculation to be 11×11, configure the weights in the matching cost calculation formula as λ1=0.85 and λ2=0.15, and set the search range for stereo matching to 192×7.

[0127] Such as Figure 1-2 As shown, S2. Input the target image and the reference image and traverse each pixel in the image, use the gray information of the 5×5 neighborhood to calculate the ZSAD information, and use the gray information of the 5×5 neighborhood to calculate the Census information...

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 rapid stereo matching method based on ZSAD-Census, and the method comprises the following steps: S1, carrying out the initialization configuration before the stereo matching of each frame of image is started; S2, inputting a target image and a reference image, traversing each pixel in the images, and calculating ZSAD information and Census information; S3, jointly using the ZSAD information and the Census information of the reference image, or adopting independent ZSAD information or Census information, constructing a hash function, constructing hash tables through the hash function, wherein each hash table is used for solving hash conflicts by constructing a chain table; and S4, calculating the hash value of the current pixel, and searching the position of the reference pixel with the same hash value as the current pixel. According to the invention, the problems of large power consumption and high area cost in the prior art are solved, and the accuracy and anti-interference performance of a stereo matching algorithm are improved.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a fast stereo matching method based on ZSAD-Census. Background technique [0002] Stereo matching is an important research content in the field of computer vision, and it is also a research hotspot in the past 30 years. Stereo matching is a technique for extracting depth information from planar images, by matching two or more images to obtain a dense disparity map, thereby establishing 3D depth information in the scene. The stereo matching algorithm includes four steps of matching cost calculation, cost aggregation, disparity calculation and disparity post-processing. The matching cost calculation is to calculate each pixel in the search area of ​​the reference image, and match the corresponding points on the target image with all disparity possibilities. cost. The matching cost is the basis of stereo matching. Designing a matching cost that is resistant to noise and in...

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): G06T7/30
CPCG06T7/30G06T2207/20032
Inventor 周开伦李朔仲亚东赵凯张斌
Owner 南京物间科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap