Unlock instant, AI-driven research and patent intelligence for your innovation.

A Single Frame Image Super-resolution Reconstruction Method Based on Group Sparse Representation

A super-resolution reconstruction and super-resolution technology, which is applied in the field of single-frame image super-resolution reconstruction based on group sparse representation, can solve the problems of affecting image quality, prone to artifacts, and not considering the structural characteristics of image slices, etc., to achieve Effects of improving quality, suppressing noise and edge artifacts

Active Publication Date: 2018-03-09
GUANGZHOU CHNAVS DIGITAL TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method overcomes the "fixed K value" problem in Chang et al.'s method and achieves good results. However, this method does not consider the structural characteristics of the image slice, and the position of non-zero coefficients in the sparse coefficients is approximately random. , causing this method to be prone to artifacts near prominent edges, affecting image quality

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 Single Frame Image Super-resolution Reconstruction Method Based on Group Sparse Representation
  • A Single Frame Image Super-resolution Reconstruction Method Based on Group Sparse Representation
  • A Single Frame Image Super-resolution Reconstruction Method Based on Group Sparse Representation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0071] Aiming at the problems of the prior art that the K value is fixed and artifacts are prone to appear near significant edges, the present invention proposes a single-frame image super-resolution reconstruction method based on group sparse representation. The group sparseness in the present invention means that the signal to be represented can be represented by a group of basis vectors with the most similar structure in the over-complete dictionary. Described in mathematical expressions as:

[0072]

[0073] arg is the English abbreviation of element (variable). The arg min f(x,t) function is the value of x and t when the following formula f(x,t) reaches the minimum value.

[0074] Among them, x is the signal to be represented, For a complete dictionary, is the i-th group in the dictionary, d ij is the j-th atom of the i-th group.

[0075] The above minimum equation can be solved by the Group Orthogonal Matching Pursuit (GOMP) algorithm.

[0076] The single-frame...

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 single-frame image super-resolution reconstruction method based on group sparse representation, including: S1, constructing a training sample library of high-resolution images, and then solving the sparse coefficient matrix of the training sample by using an orthogonal matching pursuit method, and then obtaining The group sparse dictionary of the training sample, the sparse coefficient matrix of the training sample constrains the position where the non-zero value appears in the sparse coefficient; S2, perform super-resolution image reconstruction on the low-resolution image according to the group sparse dictionary group of the training sample , to obtain a super-resolution image. The present invention considers the structural features of the image slices, uses the characteristics of group sparseness to constrain the positions where the non-zero values ​​of the sparse coefficients appear, so that the positions where the non-zero coefficients appear are no longer random, effectively suppressing the generation of noise and edge artifacts, Improved the quality of reconstructed images. The invention can be widely used in the field of image processing.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a single-frame image super-resolution reconstruction method based on group sparse representation. Background technique [0002] The goal of image super-resolution reconstruction is to obtain its high-resolution estimate by means of software calculation based on the input single-frame low-resolution image. This technology is widely used in high-definition display, video surveillance and other fields. Its implementation methods can be mainly divided into methods based on interpolation, methods based on reconstruction and methods based on learning. In recent years, with the rise of machine learning technology, learning-based methods have gradually attracted people's attention. [0003] At present, in the learning-based single-frame image super-resolution reconstruction method, Chang et al. introduced the idea of ​​manifold learning into the image super-resolution reconstruction tec...

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): G06T5/50G06T3/40
Inventor 李键红
Owner GUANGZHOU CHNAVS DIGITAL TECH