An Image Registration Method Based on Sparse Feature Matching Based on Local Linear Constraints

An image registration and local linear technology, applied in the field of image processing, can solve problems such as non-rigid deformation, achieve good robustness, improve the efficiency of image registration, and improve the effect of image registration accuracy

Active Publication Date: 2017-02-22
CHINA UNIV OF GEOSCIENCES (WUHAN)
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In order to solve the deficiencies of the prior art, the present invention provides an image registration method based on sparse feature matching based on local linear constraints. Aiming at the problem of non-rigid deformation caused by terrain fluctuations in remote sensing images, local linear constraints are applied to point matching, which can Improving image registration accuracy by preserving local structure in feature sets after image transformation

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
  • An Image Registration Method Based on Sparse Feature Matching Based on Local Linear Constraints
  • An Image Registration Method Based on Sparse Feature Matching Based on Local Linear Constraints
  • An Image Registration Method Based on Sparse Feature Matching Based on Local Linear Constraints

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The present invention will be further described below in conjunction with embodiment.

[0025] The invention provides an image registration method based on sparse feature matching of local linear constraints, comprising the following steps:

[0026] (1) Using the feature detection method to infer the hypothetical matching set of the two images to be registered x n with y n Two-bit column vectors representing the spatial positions of the feature points in the two images to be registered respectively; the hypothetical matching set S includes wrong matching and correct matching, wherein the correct matching is determined according to the geometric transformation Z between the two images with matching , that is, if (x n ,y n ) is a correct match, then y n =Z(x n ) is a correct match;

[0027] (2) The initial position plus a displacement function v will define a transformation Z: Z(x)=x+v(x), where v is simulated in a function space H that is a vector-valued regenera...

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 present invention provides an image registration method based on sparse feature matching based on local linear constraints. First, a hypothetical matching set is inferred, and then the transformation Z is defined by a displacement function, and the matching credibility is indicated by a diagonal matrix, and the weight matrix is ​​obtained by solving the weight matrix. Energy function. Finally, deterministic annealing technology is used to optimize and solve the energy function to obtain the transformation Z, and perform image registration through transformation Z combined with bilinear interpolation. The present invention provides an image registration method based on sparse feature matching based on local linear constraints. Aiming at the problem of non-rigid deformation caused by terrain fluctuations in remote sensing images, local linear constraints are applied to point matching, which can protect the feature concentration after image transformation. local structure, thereby improving image registration accuracy.

Description

technical field [0001] The invention relates to an image registration method for sparse feature matching based on local linear constraints, and belongs to the technical field of image processing. Background technique [0002] Image registration is a fundamental and challenging problem in the field of remote sensing and a prerequisite in many wide-ranging applications including terrain reconstruction, environmental monitoring, change detection, image mosaicing, image fusion, and map updating, etc. [0003] Image registration aims to establish the correspondence between pixels between two images of the same scene acquired at different times and from different perspectives or by different detectors. Registration problems can be classified as rigid or nonrigid, depending on the application and data format. Rigid registration (involving only a small number of parameters) is relatively easy and has been extensively studied. In contrast, non-rigid registration is more difficult, ...

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 Patents(China)
IPC IPC(8): G06T7/38G06T7/33
CPCG06T2207/10004
Inventor 陈珺罗林波刘超王勇罗大鹏
Owner CHINA UNIV OF GEOSCIENCES (WUHAN)
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