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

A Multimodal Medical Image Fusion Method Based on Low-rank Decomposition and Sparse Representation

A sparse representation and low-rank decomposition technology, applied in the field of image processing, can solve problems such as single function, achieve concise representation, convenient processing, and good robustness

Active Publication Date: 2020-07-10
云南联合视觉科技有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Compared with the above four types of algorithms, the other four fusion methods have relatively single functions

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
  • A Multimodal Medical Image Fusion Method Based on Low-rank Decomposition and Sparse Representation
  • A Multimodal Medical Image Fusion Method Based on Low-rank Decomposition and Sparse Representation
  • A Multimodal Medical Image Fusion Method Based on Low-rank Decomposition and Sparse Representation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] Embodiment 1: a multimodal medical image fusion method based on low-rank decomposition and sparse representation, the specific steps of the method are as follows:

[0027] Step1, two CT and MRI images with a pixel size of 256×256 to be fused (such as figure 2 , 3 shown), using matrix low-rank decomposition theory to perform low-rank decomposition to obtain low-rank partial images and sparse partial images respectively (that is, after CT image decomposition, a low-rank partial image A 1 and a sparse partial image A 2 , the MRI image is decomposed to obtain a low-rank partial image B 1 and a sparse partial image B 2 );

[0028] Step2. Use K-means singular value decomposition algorithm to select non-medical image sets (such as Figure 4 shown) to train a low-rank dictionary, and use the K-means singular value decomposition algorithm to perform low-rank decomposition on the selected non-medical image set to obtain a sparse part of the image set (such as Figure 5 sho...

Embodiment 2

[0033] Embodiment 2: a kind of multimodal medical image fusion method based on low-rank decomposition and sparse representation, the concrete steps of described method are as follows:

[0034] Step1, two MRI and PET images with a pixel size of 256×256 to be fused (such as Figure 7 , 8 shown), using matrix low-rank decomposition theory to perform low-rank decomposition to obtain low-rank partial images and sparse partial images respectively;

[0035] Step2. Use K-means singular value decomposition algorithm to select non-medical image sets (such as Figure 4 shown) to train a low-rank dictionary, and use the K-means singular value decomposition algorithm to perform low-rank decomposition on the selected non-medical image set to obtain a sparse part of the image set (such as Figure 5 shown) to train a sparse dictionary;

[0036] Step3, using sparse representation method to low-rank partial image A 1 、B 1 and the sparse part of the image A 2 、B 2 Perform sparse reconstru...

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 multimodal medical image fusion method based on low-rank decomposition and sparse representation. Firstly, low-rank decomposition is performed on two different multimodal medical images to be fused to obtain a low-rank partial image and a sparse partial image respectively. ; Use the KSVD algorithm to train a low-rank dictionary on the selected non-medical image set, and use the KSVD algorithm to perform low-rank decomposition on the selected non-medical image set to train the sparse dictionary; use the sparse representation method to train the sparse part of the image The low-rank reconstructed image and the sparsely reconstructed image are respectively obtained by sparse reconstruction with the sparse part of the image; the low-rank reconstructed image and the sparse reconstructed image are sparsely fused using the sparse representation method to obtain the fused image; two different multiple The difference between the modality medical image and the sparsely reconstructed image and the low-rank reconstructed image; the difference is added to the fused image to obtain the final sparse fused image. The present invention is superior to traditional fusion methods in both subjective and objective evaluation indicators.

Description

technical field [0001] The invention relates to a multimodal medical image fusion method based on low-rank decomposition and sparse representation, belonging to the field of image processing. Background technique [0002] In the field of image processing, image fusion is a very promising research. Image fusion technology synthesizes a fused image by synthesizing useful information of multi-sensor images of the same scene or the same sensor images of different scenes. The synthesized fusion image has all the characteristic information of the former, which is more suitable for later processing and research. An efficient fusion method can handle multi-channel information according to actual needs. These advantages make image fusion very popular in many fields. In particular, in medical imaging, ultrasound images, X-rays, computerized tomography (CT), magnetic resonance imaging (MRI), digital subtraction imaging (DSA), positron emission tomography (PET), Various modal image ...

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): G06T5/50G06K9/68G06K9/62
CPCG06T5/50G06T2207/20221G06V10/75G06F18/25
Inventor 李华锋邓志华余正涛王红斌
Owner 云南联合视觉科技有限公司
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