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

Convolutional neural network defogging algorithm based on regional division and heavy fog preprocessing

A convolutional neural network and region division technology, applied in the field of algorithm for dehazing by a method, can solve problems such as inaccurate estimation of transmittance in dense fog areas

Active Publication Date: 2017-10-27
TIANJIN UNIV
View PDF6 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] The main purpose of the present invention is to solve the problem of inaccurate estimation of the transmittance in the dense fog area existing in the existing learning-based defogging algorithm, and propose a method for distinguishing and processing thick and thin fog image blocks and predicting dense fog image blocks. Processed Image Dehazing Algorithm

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
  • Convolutional neural network defogging algorithm based on regional division and heavy fog preprocessing
  • Convolutional neural network defogging algorithm based on regional division and heavy fog preprocessing
  • Convolutional neural network defogging algorithm based on regional division and heavy fog preprocessing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0057] This patent proposes a convolutional neural network defogging algorithm based on dense and thin fog area division and dense fog area preprocessing. First, the dark channel value of the local image block in the foggy image is extracted and compared with the threshold, and the image block is judged as a dense fog image block or a hazy image block. If it is judged to be a dense fog image block, the image block will be enhanced, and then input into the convolutional neural network, and the transmittance of the image block will be estimated through the convolutional network; if it is a hazy image block, the image block will be extracted. The chromaticity feature map and dark channel feature map of the image block are input to the convolutional neural network to judge the transmittance of the image block. Finally, on the basis of the transmittance value of the image block, the original fog-free image is calculated through the imaging model of the fog image. The specific plan...

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 relates to a convolutional neural network defogging algorithm based on regional division and heavy fog preprocessing. The algorithm comprises the following steps that: dividing a foggy image into non-overlapped image blocks; for each image block, calculating the dark channel value Di of each image block; distinguishing a heavy fog image block and a mist image block; independently estimating transmissivity; and defogging Pi to obtain a fogless image block.

Description

technical field [0001] The invention relates to an algorithm for recovering image clarity in the fields of computer vision and image processing, in particular to an algorithm for defogging by using a learning method Background technique [0002] The image defogging algorithm is an algorithm for recovering the original fog-free image from the foggy image. The main purpose is to improve the clarity of the image that is affected by the fog and the image quality is deteriorated. It is widely used in transportation, satellite remote sensing, and video surveillance. , national defense and military and other industries that have high requirements for image quality. [0003] At present, many methods are committed to predicting the transmittance by learning the relationship between the characteristics reflecting the size of the fog and the transmittance under the framework of learning, and finally recovering the original fog-free image through the imaging model of the foggy image. T...

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): G06T5/00
CPCG06T2207/10024G06T2207/20021G06T2207/20084G06T2207/20081G06T5/90G06T5/73
Inventor 庞彦伟廉旭航
Owner TIANJIN 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