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

Image elastic registrating method based on limited sampling global optimisation

A global optimization, image technology, applied in image analysis, image data processing, graphics and image conversion, etc., can solve the problems of simulated annealing and genetic methods, such as long computing time, inability to guarantee global solutions, and long computing time.

Inactive Publication Date: 2007-06-06
SOUTHERN MEDICAL UNIVERSITY
View PDF0 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Among them, simulated annealing and genetic methods take too long to calculate, and cannot guarantee the final global solution; the disadvantage of tunneling method is that its optimization object has certain limitations, and its versatility is not good; There is no clear calculation method for the calculation of the number of points. To obtain a quasi-global solution, the only way to increase the number of random points is that the calculation time is too long, and the accuracy of the final registration result cannot be guaranteed.

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 elastic registrating method based on limited sampling global optimisation
  • Image elastic registrating method based on limited sampling global optimisation
  • Image elastic registrating method based on limited sampling global optimisation

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0028] Example 1 (registration of two-dimensional images):

[0029] For the needs of medical research, to estimate the motion parameters of the human heart, the inventor collected a group of MRI images of cardiac perfusion from the General Hospital of Tianjin Medical University, a total of 20, with a size of 256×256 pixels, and the selected image is the serial number The two adjacent images 5 and 3 are the reference image and the deformed image (see Figure 2a and Figure 2b). The registration test platform is a PC with WindowXP as the operating system and Matlab6.5 as the programming language. The specific registration process is as follows:

[0030] Define the deformation function and objective function before registration, the specific steps are as follows:

[0031] (1) Let the reference image and the deformed image be defined as f r (x) and f t (x), wherein the variable x is a two-dimensional vector, then the deformation function of the connection reference image and the...

example 2

[0075] Example 2 (3D image registration):

[0076] The three-dimensional image used in this embodiment is an MRI image selected from the image database of the Institute of Medical Information, School of Biomedical Engineering, Southern Medical University, and the size of the image is 128×128×128 pixels. The registration test platform is a PC with WindowXP as the operating system and Matlab6.5 as the programming language. The specific registration process is as follows:

[0077] In this embodiment, the definition method of the deformation function and the objective function is the same as Example 1, specifically:

[0078] (1) Let the reference image and the deformed image be defined as f r (x) and f t (x), where x is a three-dimensional vector. B-splines are used to construct the deformation function linking the reference image and the deformed image to obtain a parametric elastic model:

[0079] g ( x , ...

example 3

[0089] Example 3 (comparison experiment of registration accuracy):

[0090] Since the experimental effect of the method of the present invention can be better checked under the known conditions of deformation, the inventor adopts the method of artificial deformation to produce deformed images, and randomly selects 16 images from the laboratory gallery as a reference image. Among the selected reference images, there are 8 CT images and 8 MR images, and the same deformation processing is carried out on them respectively. The size of the original image is 256×256 pixels. The content of the deformation includes translation of 10 pixels along the X and Y axes, rotation of 10 degrees clockwise, and twisting elastic deformation realized by software photoshop. The reference method compared with the present invention is the elasticity based on B-splines in the article "Fast Parametric Elastic Image Registration" published by Kybic J and Unser M in "IEEE Trans. Image Processing" Volum...

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 opens an elastic image registration method based on limited sampling and global optimization. The steps of the registration are: First read the deformation image; compute the maximum frequency of defined target function, perform the average sampling calculation on the deformation coefficients with two times of the maximum frequency of target function; get the sequence of deformation images; then read into the reference images; deform the deformation images to finish the image registration with the deformation function that corresponds to minimum target function value. Furthermore, start from the coefficient value of deformation function in the rough registration; search the global minimum point in target function with the gradient descent algorithm.

Description

technical field [0001] The invention relates to graphic image conversion in an image plane, in particular to a globally optimized image elastic registration method, which is applicable to the fields of image comparison, data fusion, change analysis, target recognition and the like. Background technique [0002] Image registration refers to seeking a spatial transformation for an image so that it can achieve spatial matching with the corresponding points on another image. This kind of matching means that the points with the same content have the same spatial position in the two matching images. [0003] There are many commonly used image registration methods, but their registration steps are basically the same, mainly including the following steps: [0004] (1) Construct a deformation function that deforms the image. Existing registration methods can be divided into rigid registration methods and elastic registration methods. The deformation function used in the rigid regi...

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): G06T3/00G06T7/00
Inventor 陈武凡刘新刚
Owner SOUTHERN MEDICAL UNIVERSITY
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