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

ORB robot image registration method

An image registration and robotics technology, applied in the image field, can solve the problems of no scale invariance, the influence of image registration accuracy, and low computational complexity, so as to solve the problem of scale invariance, improve the accuracy of image registration, The effect of guaranteeing accuracy

Active Publication Date: 2020-05-29
HEFEI SIWILL INTELLIGENT +1
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Feature-based image registration has the characteristics of good robustness and low computational complexity. ORB image registration algorithm is a commonly used image registration algorithm, but it has the problem of not having scale invariance, and image registration is accurate. degrees affected

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
  • ORB robot image registration method
  • ORB robot image registration method
  • ORB robot image registration method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] refer to figure 1 , a kind of ORB robot image registration method that the present invention proposes, comprises the following steps.

[0029] S1. Extracting feature points from the first image by using a Hessian matrix to detect feature points.

[0030] refer to figure 2 , in this step, the extraction of feature points is as follows:

[0031] S11. Set the coordinates of any point P in the first image as (x, y), and define the Hessian matrix whose scale is σ as: Among them, Lxx(P, σ), Lxy(P, σ), and Lyy(P, σ) are the second derivatives of the Gaussian filtered image point P in the x direction, xy direction and y direction, respectively.

[0032] In the specific implementation of this step, since the Gaussian filter is required first to obtain the Hessian, and then the second order derivative is obtained, in the discrete pixel points, a template can be used instead, such as Figure 4 shown. This template can greatly increase calculation speed. Figure 4 The top r...

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 an ORB robot image registration method. The method comprises the following steps: extracting feature points from a first image by using a method of detecting the feature pointsby using a Hessian matrix; for each feature point, obtaining a quotient obtained by dividing the distance between the feature point and the corresponding suboptimal matching point by the distance between the feature point and the corresponding optimal matching point as a measured value; removing the feature points of which the measured values are smaller than a preset first threshold value; detecting the remaining feature points by adopting a symmetric test, reserving the symmetric feature points and the corresponding optimal matching points as matching pairs, and collecting all the matching pairs as samples; obtaining a final mapping model from the feature points to the optimal matching points according to sample training; and for two images to be registered, inputting the feature pointsextracted from one image into the final mapping model, and outputting matching points corresponding to the feature points on the other image by the final mapping model. According to the invention, while the problem of scale invariance is solved, the image registration precision is greatly improved.

Description

technical field [0001] The invention relates to the field of image technology, in particular to an ORB robot image registration method. Background technique [0002] Image registration is the basic problem of robot image processing. Image registration algorithm can match two different images, and then obtain the transformation relationship between them. A prerequisite for applications such as 3D vision. Feature-based image registration has the characteristics of good robustness and low computational complexity. ORB image registration algorithm is a commonly used image registration algorithm, but it has the problem of not having scale invariance, and the image registration is accurate. degrees are affected. Contents of the invention [0003] Based on the technical problems existing in the background technology, the present invention proposes an ORB robot image registration method. [0004] A kind of ORB robot image registration method that the present invention proposes,...

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): G06K9/00G06K9/62G06T7/33
CPCG06T7/33G06V20/10G06V10/757Y02T10/40
Inventor 盛恩菊郑博超陈静朱玉王英明李琳琳
Owner HEFEI SIWILL INTELLIGENT
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