Form and color fused image super-resolution reconstruction model construction and reconstruction method

A technology of super-resolution reconstruction and construction method, which is applied in the field of image super-resolution reconstruction model construction and reconstruction that integrates shape and color, and can solve problems such as excessively smooth texture, blurred structure boundaries, and chaotic brightness and color.

Active Publication Date: 2019-10-22
NORTHWEST UNIV(CN)
View PDF7 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide an image super-resolution reconstruction model construction and reconstruction method that integrates shape and color, so as to solve the problems of super-resolution images obtained in the prior art, such as excessively smooth texture, blurred and unprotrusive structure boundaries, and local areas. Problems such as uniform distribution of internal brightness and color confusion

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
  • Form and color fused image super-resolution reconstruction model construction and reconstruction method
  • Form and color fused image super-resolution reconstruction model construction and reconstruction method
  • Form and color fused image super-resolution reconstruction model construction and reconstruction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0063] In this embodiment, a method for constructing an image super-resolution reconstruction model that fuses shape and color is disclosed, such as figure 1 shown, follow the steps below:

[0064] Step 1. Obtain multiple high-resolution images and obtain the target image set X, where X={X 1 ,X 2 ,...,X n ,...,X N},X n is the nth high-resolution image, N is a positive integer, n∈N;

[0065] Downsampling is performed on each high-resolution image to obtain multiple low-resolution images, and an image set x to be reconstructed is obtained, where x={x 1 ,x 2 ,...,x n ,...,x N}, x n is the nth low-resolution image;

[0066] In this embodiment, the model provided by the present invention is trained and tested using the training set and the test set, wherein the 1-800 high-resolution images in the DIV2K data set are down-sampled by a ratio of 4 to generate the corresponding low-resolution images images to construct a training dataset consisting of high / low resolution imag...

Embodiment 2

[0145] This embodiment discloses an image super-resolution reconstruction method that combines shape and color. After the image to be reconstructed is input into the image super-resolution reconstruction model described in Embodiment 1, a reconstructed image is obtained.

[0146] The following is the analysis of the experimental results of this experiment:

[0147] The comparative methods used in this experiment include: Bicubic, DRCN, VDSR, SRResNet, SRGAN-MSE, SRGAN, and EnhanceNet methods. figure 2 The SR reconstruction result of a low-resolution image is given, where (a) is the original HR image, (b) is the HR image obtained after bicubic interpolation of the LR image, (c) is the result of DRCN reconstruction, (d) is the result of VDSR, (e) is the reconstruction result of SRGAN, (f) is the reconstruction result of EnhanceNet, and (g) is the reconstruction result of the method of the present invention. As can be seen from the result figure, the super-resolution images obt...

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 form and color fused image super-resolution reconstruction model construction and reconstruction method. Only an image pixel loss optimization super-resolution model is used,so that the balanced distribution of the structure, texture and color distribution in the image can be ignored; the phenomena of unsmooth structure, unclear texture, unsmooth color, excessive rendering dislocation and the like of a super-resolution reconstruction result are caused; the super-resolution reconstruction method for constructing the image adversarial fusion morphological component adversarial network and the color adversarial network solves the problem that the structure, texture and color image characteristics cannot be considered in the prior art, and effectively improves the reconstruction quality of the image.

Description

technical field [0001] The invention relates to an image super-resolution reconstruction method, in particular to an image super-resolution reconstruction model construction and reconstruction method that fuses shape and color. Background technique [0002] Scientific research and statistics show that about 75% of the information that humans obtain from the outside world comes from images. In the process of image acquisition, due to the limitation of imaging distance, imaging device resolution and other factors, the imaging system is usually affected by degradation factors such as deformation, blur, downsampling and noise, and cannot accurately obtain the information in the original scene. Super-resolution (SR) reconstruction technology is to improve the resolution of images through software methods, which can fully use the existing data to improve the visual effect of images at low cost. This technique uses the information of one or more low-resolution images to recover th...

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 Applications(China)
IPC IPC(8): G06T3/40
CPCG06T3/4053G06T3/4046
Inventor 祝轩程悦孙逸霏彭进业章勇勤张超王荣智宋奇衡柳欣
Owner NORTHWEST UNIV(CN)
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