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

Characteristic learning-based single image defogging method

A feature learning, single image technology, applied in the field of computer vision, can solve the problem of transmittance estimation accuracy limitation, not considering texture structure features, etc.

Active Publication Date: 2016-06-29
SOUTH CHINA AGRI UNIV
View PDF2 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is a common limitation in these two types of methods, that is, only the color features are used, and the texture structure features related to the haze are not considered, which limits the estimation accuracy of the transmittance, so it has certain scene limitations. sex

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
  • Characteristic learning-based single image defogging method
  • Characteristic learning-based single image defogging method
  • Characteristic learning-based single image defogging method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] The present invention will be further described below in conjunction with the accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0050] The schematic diagram of the execution steps of the method of the present invention is as attached image 3 As shown, it specifically includes the following steps:

[0051] S1. Acquire a set of foggy images I set and its corresponding scene depth map d set as a training dataset;

[0052] S2. Using Dataset I set and d set Training a sparse autoencoder for extracting fog-related texture-structural features. figure 1 A sparse autoencoder and its training process for extracting scene texture and structure features are given. The autoencoder consists of two parts: the first part is sparse coding, which is used to extract the main texture and structural features of local blocks, and is realized by an unsupervised self-learning neural network, such as figure 1 As shown in (c), the input of the...

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 characteristic learning-based single image defogging method. Firstly, multi-scale extraction of texture structure characteristics of a foggy image is performed via a sparse autoencoder, and at the same time, various color characteristics related with fog are extracted. Then, a multi-layer neural network is adopted for sample training, the mapping relationship between the texture structure characteristics and the color characteristics and the scene depth in the foggy condition is learned, and a scene depth chart of the foggy image is estimated. On this basis, a transmittance chart is approximately estimated by using the scene depth chart. The transmittance chart effectively reflects the fog concentration of each local area in the foggy image. Finally, with the combination of an atmospheric scattering model, restoration is further carried out according to the transmittance chart and a fogless image is obtained. The invention allows restoration of foggy images so as to obtain high-quality fogless images. In addition, compared with conventional defogging methods, the characteristic learning-based single image defogging method of the invention achieves better universal scene adaptability.

Description

technical field [0001] The present invention relates to the field of computer vision, and more specifically, to a single image defogging method based on feature learning. Background technique [0002] In severe weather such as fog and haze, there are many water droplets or dust particles in the atmosphere, causing light to be scattered or absorbed during transmission. At this time, images taken outdoors are often blurred and have low contrast. Systems that depend on outdoor imagery, such as surveillance equipment, experience a sharp drop in performance or fail to function properly. Therefore, it is of great significance to dehaze the image. [0003] At present, the methods of image defogging can be mainly divided into two categories: one is the method of image enhancement, that is, the purpose of defogging is achieved by enhancing the contrast of the image. This kind of method can be applied and targeted to improve the existing mature images. The processing algorithm can 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
CPCG06T2207/20084G06T2207/20081G06T5/73
Inventor 王美华梁云麦嘉铭
Owner SOUTH CHINA AGRI 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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com