Image fusion and super-resolution joint implementation method based on discriminant dictionary learning

A dictionary learning and fusion method technology, applied in the field of digital image processing, can solve problems such as reducing visual quality

Active Publication Date: 2020-10-16
KUNMING UNIV OF SCI & TECH
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this approach is likely to introduce artifacts created in the fir

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 fusion and super-resolution joint implementation method based on discriminant dictionary learning
  • Image fusion and super-resolution joint implementation method based on discriminant dictionary learning
  • Image fusion and super-resolution joint implementation method based on discriminant dictionary learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0094] Example 1: Such as Figure 1-6 Said, a multi-source image fusion method based on discriminative dictionary learning and morphological component decomposition, includes the following steps:

[0095] 1) Construct training samples for dictionary learning. Select 8 HR training samples to train the high-resolution image discriminant dictionary pair, then down-sampling, and then up-sampling through bicubic interpolation to restore the same size as the high-resolution image as the corresponding LR training image;

[0096] 2) Generate an initial dictionary randomly. Then a new dictionary learning method is proposed to decompose the input image into low-rank and sparse components. In order to realize image fusion and super-resolution reconstruction simultaneously, a pair of HR dictionaries and a pair of LR dictionaries are jointly trained, as well as the coefficient conversion matrix H between the coding coefficients of high and low resolution image blocks. A pair of HR dictionaries...

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 relates to an image fusion and super-resolution joint implementation method based on discriminant dictionary learning, and belongs to the technical field of digital image processing. Specifically, firstly, two pairs of low-rank and sparse dictionaries and a high-resolution and low-resolution image coding coefficient conversion matrix are jointly trained; wherein one pair of dictionaries is used for representing low-rank and sparse parts of an input image, the other pair of dictionaries is used for reconstructing high-resolution fused low-rank and sparse parts, and the conversionmatrix is used for establishing a potential relationship between a high-resolution image and a low-resolution image. And then a sparse and low-rank separation model is constructed, and the input imageis effectively decomposed into low-rank and sparse parts, so that a high-resolution fusion image can be constructed through different dictionaries. According to the invention, image fusion and super-resolution reconstruction are realized in a combined manner. Experimental results show that the method has better fusion performance no matter in the aspects of visual effect or objective index.

Description

Technical field [0001] The invention relates to a joint realization method of image fusion and super-resolution based on discriminant dictionary learning, and belongs to the technical field of digital image processing. Background technique [0002] Image fusion can integrate complementary information about the same scene acquired by different sensors into one image, and can provide a more comprehensive and accurate description of the scene, thereby helping to identify events and objects. In recent years, this technology has attracted more and more attention from researchers, and significant research progress has been made. [0003] Existing image fusion methods can be roughly divided into three categories, namely methods based on multiscale transform (MST), methods based on dictionary learning (DL) and methods based on deep learning. In MST-based methods, commonly used MSTs include wavelet transform, dual tree complex transform (DTCWT), shear transform, curvelet transform, contour...

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): G06T3/40G06T5/50
CPCG06T3/4053G06T5/50G06T2207/20081G06T2207/20221
Inventor 李华锋陈怡文杨默远余正涛张亚飞
Owner KUNMING UNIV OF SCI & 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
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap