Image registration method and device

An image registration and image technology, applied in the field of image processing, can solve the problems of low image registration accuracy, poor effect, unstable feature descriptors, etc., and achieve the effect of improving the effect, high registration accuracy and high stability

Active Publication Date: 2017-07-14
BEIJING UNIV OF POSTS & TELECOMM
View PDF5 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, there is generally no linear correspondence between the corresponding pixel values ​​in different spectral images of the same scene, and their texture structure information is also very different, which makes the feature descriptors determined by the above method unstable and makes the accuracy of image registration lower, less effective

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
  • Image registration method and device
  • Image registration method and device
  • Image registration method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] The invention provides an image registration method. figure 1 Flow chart of an image registration method provided by the present invention Figure 1 . Such as figure 1 As shown, the image registration method may include:

[0039] S101. Perform affine transformation (AffineTransform) on the image to be registered according to preset reference feature points to obtain an affine transformed image.

[0040] Specifically, the preset reference feature points may include: at least one of reference image features such as pixel point features, straight line features, and curve features.

[0041] The image to be registered may be a multispectral image. For image registration, a reference image is required for the image to be registered. The reference image can also be a multispectral image. The reference image and the image to be registered may be any two multispectral images in the plurality of multispectral images. The multiple multispectral images may be multiple images...

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 image registration method and device. The image registration method may comprises steps of: subjecting a to-be-registered image to affine transformation according to a preset reference feature point to obtain an image subjected to the affine transformation; processing a reference image and the image subjected to the affine transformation by using a antoencoding algorithm to obtain an image feature; performing CNN training based on the image feature, preset matched image data and preset unmatched image data to obtain a feature descriptor; and subjecting the reference image and the to-be-registered image to image registration according to the feature descriptor. The image registration method and device achieve high registration accuracy of multispectral images and improve an image registration effect.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image registration method and device. Background technique [0002] Image registration is the process of matching multiple images of the same scene from different times, different sensors, different illuminance or different viewing angles, and transforming the multiple images into the same coordinate system for There are splicing and fusion between images such as offset, rotation, and proportion. Multi-spectral image refers to multiple images of the same scene obtained according to different spectra. To register multi-spectral images, the temporal and spatial correlation and information complementarity of multiple images can be used to obtain a comprehensive and comprehensive view of the scene. Clear description, which in turn increases spatial resolution, reduces blurriness, and enhances images. [0003] The line feature matching method, as the most common ...

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): G06T3/00G06N3/08
CPCG06N3/08G06T3/0075
Inventor 李永刘杰魏丽君何欣徐良鹏
Owner BEIJING UNIV OF POSTS & TELECOMM
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