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

Image super-resolution method based on sparse regularization technology and weighted guidance filtering

A guided filtering and super-resolution technology, applied in the field of learning-based super-resolution, can solve the problem of insufficient information recovery such as edges, textures and structures

Active Publication Date: 2018-01-19
HUAQIAO UNIVERSITY
View PDF5 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of this, the object of the present invention is to provide an image super-resolution method based on sparse regularization technology and weighted guided filtering, to overcome the problem of insufficient information restoration of existing methods such as edges, textures and structures, and further improve the quality of reconstructed images. 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
  • Image super-resolution method based on sparse regularization technology and weighted guidance filtering
  • Image super-resolution method based on sparse regularization technology and weighted guidance filtering
  • Image super-resolution method based on sparse regularization technology and weighted guidance filtering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0070] Such as figure 1 The image super-resolution method disclosed in this embodiment based on sparse regularization technology and weighted guided filtering specifically includes the following steps:

[0071] S1: Input LR image Y to be reconstructed, HR image training set TI h , first to TI h The sample image in is down-sampled to get the LR sample image set TI l . The downsampling model used is TIl l =DBTI h +n, where D is the downsampling operator, B is the fuzzy matrix, n is random additive noise, and then TI h and TI l Using the joint dictionary training algorithm to get the HR dictionary Φ h and the LR dictionary Φ l Then by FSS ((Feature sign search, feature representation search) algorithm solving traditional sparse coding objective function shown in formula (1), obtain the sparse representation coefficient α corresponding to Y, described formula (1) is as follows:

[0072]

[0073] where ||α|| 0 Indicates the number of non-zero values ​​contained in the ...

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 discloses an image super-resolution method based on the sparse regularization technology and the weighted guidance filtering. The method is characterized in that: a new sparse coding objective function is constructed by combining the non-local similarity of the image and the manifold learning theory, so that on one hand, similar image blocks are searched in the initial reconstruction image to construct the no-local similarity regularization term, and non-local redundancy of the image is obtained to maintain the edge information, and on the other hand, the local linear embedding method is used to construct the manifold learning regularization term, and the prior knowledge of the structure of the image is obtained to enhance the structure information; and the global error compensation model of the weighted guidance filtering is used to carry out error compensation on the reconstructed high-resolution image to obtain the image with a smaller reconstruction error andhigher quality.

Description

technical field [0001] The invention relates to a learning-based super-resolution method, in particular to a super-resolution method based on sparse regularization technology and weighted guided filtering. Background technique [0002] The spatial resolution of an image is an important indicator of image quality. Generally, the higher the spatial resolution of an image, the richer the details of the image and the stronger its ability to express information, which is more conducive to subsequent image processing and analysis. and understanding. At present, in medical diagnosis, pattern recognition, video surveillance, biometric identification, remote sensing imaging and other application fields, image processing systems often need to obtain high-resolution images to improve the reliability of analysis results. However, in practical applications, due to the limitation of the physical resolution of the imaging system and the influence of many factors such as scene changes and ...

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): G06T3/40G06T5/00G06K9/62
Inventor 黄炜钦黄德天顾培婷林炎明
Owner HUAQIAO 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