Single image defogging method based on multi-scale self-attention generative adversarial network

A single image, multi-scale technology, applied in the field of image processing, can solve the problem of poor quality of dehazing images

Active Publication Date: 2021-01-05
XIAN UNIV OF TECH
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to provide a single image defogging method based on multi-scale self-attention generative confrontation network, which solves the problem of poor quality of defogged images in the prior art

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
  • Single image defogging method based on multi-scale self-attention generative adversarial network
  • Single image defogging method based on multi-scale self-attention generative adversarial network
  • Single image defogging method based on multi-scale self-attention generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0059] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0060] The present invention is based on a multi-scale self-attention generative adversarial network single image dehazing method. The input of the generator network includes three different image scales. For two branch networks, each pixel is considered in relation to all other pixels by adopting a self-attention mechanism The non-local enhanced features are calculated by the relationship, and the obtained enhanced features are input into the backbone network to enhance the image dehazing ability of the backbone network. At present, the storage location of the self-attention mechanism is generally at the front end of the network, and the calculation of attention is complicated, so the common practice is to down-sample the feature map in the backbone network and then introduce the self-attention mechanism, but this method cannot be directly 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 a single image defogging method based on a multi-scale self-attention generative adversarial network, and the method comprises the steps: carrying out the training of a generative adversarial network model constructed through the downsampling of an image twice through a training set formed by the normalization of the image, and obtaining a trained generative adversarial network model; and optimizing the defogging result by using a loss function in the training process, and finally inputting the foggy image into the generative adversarial network model to obtain a defogged image. According to the single image defogging method provided by the invention, the problem of poor defogged image quality in the prior art is solved.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to a method for defogging a single image based on a multi-scale self-attention generation confrontation network. Background technique [0002] The purpose of single image dehazing is to recover clean images from hazy images, which is essential for subsequent high-level tasks such as object recognition and scene understanding. Therefore, image dehazing has received extensive attention in the field of computer vision. According to the physical model, the image defogging process can be expressed as [0003] I(x)=J(x)t(x)+A(1-t(x)) (1); [0004] where I(x) and J(x) denote foggy and clear images, respectively. A represents the global atmospheric light, and t(x) represents the transmission map. The transmission map can be expressed as t(x)=e -βd(x) , d(x) and β represent depth of field and atmospheric scattering coefficient, respectively. Define a foggy image I(x), and most a...

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): G06T5/00
CPCG06T5/003G06T2207/20081G06T2207/20084
Inventor 石争浩周昭润
Owner XIAN UNIV OF 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