Image defogging method for haze concentration of adaptive neural network based on end-to-end

A haze concentration, neural network technology, applied in the field of image dehazing based on end-to-end haze concentration adaptive neural network, can solve the problems of complex, prior information invalidation calculation and so on

Active Publication Date: 2020-11-10
WENZHOU UNIVERSITY
View PDF7 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, different priors depend on the estimation of a certain feature of the image. In real scenes, these prior information are often invalid and computationally complex.

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 defogging method for haze concentration of adaptive neural network based on end-to-end
  • Image defogging method for haze concentration of adaptive neural network based on end-to-end
  • Image defogging method for haze concentration of adaptive neural network based on end-to-end

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] see figure 1 , figure 2 , image 3 and Figure 4 , an end-to-end haze concentration-based self-adaptive neural network image defogging method disclosed by the present invention comprises the following steps:

[0044]S1. Constructing an image dehazing model, wherein the image dehazing model includes a pyramid feature extractor, a feature enhancement module and a multi-scale feature attention module;

[0045] The specific process is as figure 2 As shown, the image dehazing model is constructed. Image dehazing models include pyramidal feature extractors such as figure 2 shown), feature enhancement modules (such as figure 2 shown) and a multi-scale feature attention module (such as figure 2 shown);

[0046] S2. Obtain foggy image data, use the pyramid feature extractor to extract the feature maps of the four different layers of the fog image through the convolutional neural network, and fuse information of different scales to generate useful information; mainly...

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 defogging method for haze concentration of an adaptive neural network based on end-to-end. The method comprises the following steps: constructing an image defogging model; acquiring foggy image data; using a feature enhancement module in the image defogging model for cascading the feature map with images recovered by different paths, combining fuzzy images with different dense haze degrees to help the network sense image haze concentration in a self-adaptive mode; reconstructing the features after function enhancement into a clear fog-free image through a multi-scale feature attention module; calculating mean square errors and perception losses of the restored images and the corresponding clear images, and updating an image defogging model; wherein the meansquare errors guide the image defogging model to learn the content of the clear images, the perception loss is used for quantizing the visual difference between the recovered images and the corresponding clear images, and the two loss functions cooperatively optimize the defogging model. According to the technical scheme, effective defogging processing is carried out on the actually shot fog image, a high-quality image is recovered, and the practicability is good.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an image defogging method based on an end-to-end haze concentration self-adaptive neural network. Background technique [0002] With the development of technology, some computer vision tasks such as object detection, object tracking, behavior analysis, face recognition, etc. have made great breakthroughs. However, advanced vision tasks such as detection and tracking rely on clear video and image data, and their performance is often greatly affected in actual scenes such as heavy fog and heavy rain. As a pre-task of some advanced vision tasks, image dehazing has attracted the attention of many researchers in recent years. [0003] Image dehazing is a typical image restoration problem, which can be traced back to 1924. McCartney et al. first proposed a classic atmospheric light scattering model. The model is as follows: [0004] I=tJ+A(1-t), [0005] t(x)=e βd(x) , ...

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): G06T5/00G06N3/04G06N3/08
CPCG06T5/003G06N3/08G06N3/045
Inventor 张笑钦王涛徐曰旺赵丽
Owner WENZHOU UNIVERSITY
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