Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

A Color Image Reconstruction Method Based on Local Data Block Tensor Enhancement

A local data and color image technology, applied in the field of image processing, can solve the problems of affecting the visual effect of the image, disrupting the data structure, and unable to reconstruct the image, etc., to achieve the effect of increasing the low-rank structure, increasing the data dimension, and improving the reconstruction effect

Active Publication Date: 2022-08-09
HEBEI UNIV OF TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although tensor enhancement technology can increase the data dimension and greatly improve the accuracy of image restoration, the current addressing mode based on this technology completely disrupts the data structure for each pixel. Serious artifact blocks will be generated, seriously affecting the visual effect of the image, and image reconstruction cannot be achieved with high accuracy

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 Color Image Reconstruction Method Based on Local Data Block Tensor Enhancement
  • A Color Image Reconstruction Method Based on Local Data Block Tensor Enhancement
  • A Color Image Reconstruction Method Based on Local Data Block Tensor Enhancement

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The technical solutions of the present invention will be described in detail below with reference to the specific drawings and embodiments, which are not intended to limit the protection scope of the present application.

[0033] The present invention is a color image reconstruction method based on local data block tensor enhancement technology (referred to as method, see Figure 1-4 ), the method comprises the following steps:

[0034] Step 1: The image to be reconstructed is a single size of 2 p ×2 p Take the standard color image of ×3 as an example, denoted as is the field of real numbers, and p is a positive integer; The image after randomly missing part of the data is For the random missing operator Ω acts on obtained from above, The gray value at the missing data is 0, and the known data is in The position in is the index position;

[0035] The local data block is used as the tensor enhancement unit for structured addressing, the pixels of the imag...

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 is a color image reconstruction method based on local data block tensor enhancement technology, comprising step 1: dividing an image to be reconstructed into local data blocks, using the local data blocks as tensor enhancement units to perform structured addressing, The image is converted from three-dimensional data to high-dimensional data, and high-order tensors are obtained. Step 2: Use the tensor chain kernel norm minimization model to reconstruct the high-order tensors to obtain the reconstructed tensors. Step 3: Convert the reconstructed tensors Perform the inverse operation of step 2 according to the index position to obtain the converted image; calculate the gray value of each pixel of the converted image, restore the converted image size to the original size of the image to be reconstructed, and complete the color image reconstruction. The method uses the local data block as a tensor enhancement unit for structured addressing, operates the local data block as a whole, retains the complete structure of the local data, and reduces the artifact blocks caused by completely disrupting the data structure on the reconstructed image.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a color image reconstruction method based on local data block tensor enhancement technology. Background technique [0002] As digital image processing is widely used in communication, medicine, aerospace and other fields, as an important research field of digital image processing, image restoration has gradually become a research hotspot. Since color images are in natural tensor form, the reconstruction problem of color images can be regarded as a tensor completion problem. [0003] The Tensor Train (TT) decomposition model solves the problem that the rank minimization scheme is inefficient in capturing the global information of tensors due to the unbalanced matrix size of the traditional tensor decomposition model by virtue of a more balanced matrixing method. The tensor chain decomposition model can The correlation between data of different dimensions is fully capture...

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): G06T5/00
CPCG06T5/007G06T2207/10004G06T2207/10024
Inventor 何静飞郑绪南高鹏周亚同
Owner HEBEI UNIV OF TECH
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
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