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

Image super-resolution method based on multi-scale attention convolutional neural network

A convolutional neural network and super-resolution technology, applied in the field of image super-resolution based on multi-scale attention convolutional neural network, can solve problems such as poor results, avoid gradient instability, and improve training and use. Speed, optimized transfer flow effect

Active Publication Date: 2019-11-05
WUHAN UNIV
View PDF3 Cites 32 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the image super-resolution algorithm is simply regarded as using a neural network to learn the mapping relationship between a low-resolution image and a high-resolution image, and the effect is not good.

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 multi-scale attention convolutional neural network
  • Image super-resolution method based on multi-scale attention convolutional neural network
  • Image super-resolution method based on multi-scale attention convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The invention discloses an image super-resolution method based on a multi-scale attention convolutional neural network, which utilizes the advantages of multi-scale units, attention units, dense connection structures, residual structures, and sub-pixel convolution layers, and can efficiently High-resolution images are reconstructed. Method of the present invention specifically comprises the following steps:

[0036] Step 1, make training set (I LR h HR );

[0037] The present invention uses the DIV2K data set commonly used in the field of image super-resolution reconstruction. The DIV2K data set contains L high-quality images of 2K resolution, including M training sets, N testing sets, and P verification sets, which contain rich The scene can be used to train and test the model. Specifically, first, 16 images are randomly sampled from the DIV2K training dataset, including low-resolution images and corresponding high-resolution images. Then, randomly select image b...

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 super-resolution method based on a multi-scale attention convolutional neural network. By using a multi-scale structure and an attention mechanism, the convolutional neural network can extract rich information in an image, and importance distinguishing can be performed on extracted features, so that important information and interference information can be distinguished in learning of the convolutional neural network, and the learning efficiency is improved. A dense connection mode and a residual connection mode are used, so that the back propagation of the gradient is enhanced, the problem of instability of the gradient is avoided, the reusability of the features is enhanced, the features in the low-resolution image are fully utilized, the training speed of the model is increased, and the parameters of the model are further reduced. Besides, a sub-pixel convolution layer method is used in an image amplification stage, learned pixel values are filled into a high-resolution image according to a certain rule without calculation, so that the operation is concentrated in a small-scale stage, the operation amount is reduced, and the image reconstructionspeed is improved.

Description

technical field [0001] The invention belongs to the field of computer vision, in particular to an image super-resolution method based on a multi-scale attention convolutional neural network. Background technique [0002] Nowadays, people's living standards are improving day by day. Mobile phones, video cameras, cameras and other electronic devices are very common in daily life, playing the role of recording and sharing beautiful moments. In addition, in traffic safety, public security criminal investigation, medical imaging, remote sensing satellite, military reconnaissance and other public fields, image and video data are also very important, and the quality of image and video data often plays a vital role. At the same time, in some other computer vision tasks, high-quality images and videos help to improve the performance of the task, which can be clearly seen from multi-task learning. [0003] As an important information carrier, images account for 70% of people's abilit...

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): G06T3/40G06T7/00G06N3/08
CPCG06T3/4046G06T3/4076G06T7/0002G06N3/084G06T2207/30168
Inventor 邹华肖田雨肖春霞姚江军
Owner WUHAN UNIV
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