Image mixing super-resolution method based on adaptive texture distillation

A super-resolution, image mixing technology, applied in the field of digital images, which can solve problems such as inability to use the same modules, enhance low-mixed resolution images, etc.

Active Publication Date: 2021-06-01
WUHAN UNIV
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Since texture enhancement and texture smoothing are opposite operations, the same modules cannot be used
Existing lightweight methods do not do well in enhancing low-mix resolution images with aliasing of real and fake textures

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 mixing super-resolution method based on adaptive texture distillation
  • Image mixing super-resolution method based on adaptive texture distillation
  • Image mixing super-resolution method based on adaptive texture distillation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031]In order to facilitate those of ordinary skill in the art to understand and implement the present invention, the present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the implementation examples described here are only used to illustrate and explain the present invention, and are not intended to limit this invention.

[0032] please see figure 1 , a kind of image mixing super-resolution method based on adaptive texture distillation provided by the present invention, comprises the following steps:

[0033] Step 1: Build a dataset;

[0034] Perform data preprocessing on the images in the original high mixed-resolution dataset, reduce the mixed resolution of the image, and construct a low / high mixed-resolution image pair dataset required for training the adaptive texture distillation network, including training data and test data; Wherein, the low mixed resolution refers to t...

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 an image mixing super-resolution method based on adaptive texture distillation. The image mixing super-resolution method mainly comprises the following steps: constructing a low/high mixing resolution image pair data set required for training a neural network; constructing a required self-adaptive texture distillation network; training and testing the adaptive texture distillation network by using the data set to obtain a trained neural network model; and reconstructing a mixed super-resolution image from the low-mixed-resolution image by using the trained network model. According to the method, the low-mixed-resolution image is used, so that the data volume is saved, and more image details can be reserved; the self-adaptive texture distillation network processes true and false textures respectively, and a mode of weighting multiple weights is adopted to make full use of features of different levels, so that the quality of a super-resolution reconstructed image is better.

Description

technical field [0001] The invention belongs to the technical field of digital images and relates to an image super-resolution method, in particular to an image hybrid super-resolution method based on adaptive texture distillation. Background technique [0002] Storing or transmitting high-resolution images requires a large amount of data, and the storage space or transmission bandwidth can be saved by reducing the image resolution. However, large-scale spatial downsampling will cause the image to lose a lot of spatial details, which is not conducive to super-resolution reconstruction. Therefore, instead of changing the spatial resolution of the image alone, the mixed resolution of the image can be reduced, that is, by using small-scale spatial downsampling with gray quantization instead of large-scale spatial downsampling, it can be guaranteed that under a similar amount of data, Preserve more spatial details of the image and enhance the quality of the super-resolution rec...

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): G06T3/40G06N3/04
CPCG06T3/4053G06T3/4046G06T3/4007G06N3/045
Inventor 韩镇刘春雷温佳兴胡辉王中元涂卫平
Owner WUHAN UNIV
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