Quantization method for calculating distribution credibility of feature matching points

A feature matching and quantization method technology, applied in the field of image processing, can solve problems that affect the effect of 3D point cloud reconstruction of images

Active Publication Date: 2019-07-19
CHENGDU UNIV OF INFORMATION TECH
View PDF6 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The distribution of feature matching points may affect the effect of image 3D point cloud reconstruction

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
  • Quantization method for calculating distribution credibility of feature matching points
  • Quantization method for calculating distribution credibility of feature matching points
  • Quantization method for calculating distribution credibility of feature matching points

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0053]The present invention introduces the horizontal precision attenuation factor (HDOP) in the GPS satellite navigation and positioning system, refers to the principle that the HDOP value estimated by evenly distributed feature matching points is the best, and constructs a quantification method for the distribution reliability of feature matching points.

[0054] Such as image 3 As shown, the quantitative method for calculating the reliability of feature matching point distribution includes the following steps: Step 1, on the basis of calculating the homography matrix of the stereo pair based on the set of feature matching points, estimating the overlapping area range and overlapping area of ​​the stereo pair Center point coordinates.

[0055] If both the left and right images of the stereo pair are understood as polygons, the calculation ...

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 quantization method for calculating distribution credibility of feature matching points, which comprises the following steps of: acquiring image plane coordinates of all feature matching points in a stereo image pair, and estimating a range of a feature matching point set in a stereo image overlapping region and central point position coordinates of the overlapping region; redistributing the feature matching points of the stereo image pair overlapping area to enable the feature matching points to be uniformly distributed; respectively calculating HDOP values corresponding to the original characteristic matching point set and the characteristic matching point set after redistribution in the stereo image pair overlapping area, namely an HDOP actual measurement theory and an HDOP theory; on the basis of calculating the relative error (a = (HDOP actual measurement-HDOP theory)/HDOP actual measurement) of the HDOP actual measurement relative to the HDOP theory, using a credibility value ((1-a) * 100%) to represent the influence degree of the extracted characteristic matching point distribution on the quality of the reconstructed three-dimensional model based onthe stereo image pair. According to the method, the internal projection relation of the image can be described for the basic matrix, and a measurement standard is provided for measuring the three-dimensional modeling precision of the image.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a quantification method for calculating the reliability of distribution of feature matching points. Background technique [0002] With the popularization and use of camera equipment such as digital cameras and mobile phones, and the resolution of camera components is getting higher and higher, the spatial information contained in images is becoming more and more abundant. Rapidly extracting and building 3D models from images has become the development of 3DGIS spatial data acquisition. trend. At present, the image-based 3D point cloud reconstruction process mainly includes feature matching, fundamental matrix estimation, camera self-calibration, and 3D point cloud calculation. Among them, feature matching is the data source for basic matrix estimation, camera self-calibration and 3D point cloud calculation. The uncertainty of feature matching points directly affects th...

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): G06K9/62
CPCG06V10/757
Inventor 卞玉霞刘学军王美珍王丽褚永彬周业
Owner CHENGDU UNIV OF INFORMATION TECH
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