Unlock instant, AI-driven research and patent intelligence for your innovation.

A Single Image Dehazing Method Based on Deep Learning

A single image, image technology, applied in the field of single image dehazing based on deep learning, to achieve the effect of improved peak signal-to-noise ratio, fast processing speed, and good dehazing effect

Active Publication Date: 2020-06-26
PEKING UNIV SHENZHEN GRADUATE SCHOOL
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a new image defogging method to solve the problem of restoring aerial video images in haze weather to high-definition fog-free images

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
  • A Single Image Dehazing Method Based on Deep Learning
  • A Single Image Dehazing Method Based on Deep Learning
  • A Single Image Dehazing Method Based on Deep Learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] Below in conjunction with accompanying drawing, further describe the present invention through embodiment, but do not limit the scope of the present invention in any way.

[0024] The present invention takes the video obtained by drone aerial photography in foggy weather as an example. First, the image of each frame in the video is taken out, and the RGB image with a resolution of 3840×2160 is used as the input of the method of the present invention, and the method will automatically extract the R channel in R, G, and B to form a two-dimensional grayscale image. The specific treatment is as follows (such as figure 1 ):

[0025] A. The scattering effect is eliminated, and the The implementation method is:

[0026] A1. The input data of this step is represented by I, and the convolution Conv1 operation is performed, where Conv1(I)=W 1 I+B 1 ;W 1 is the parameter of the convolutional layer neuron, B 1 is the offset.

[0027] A2. Perform the ReLU operation on 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 present invention discloses a single image defogging method based on deep learning, which belongs to the field of image processing. According to the formula of the fogging model I(x,y)=T(x,y)J(x,y)+(1-T(x,y))A, the present invention can transform and derive the formula using deep convolutional neural Network technology is used to obtain fog-free high-definition image J(x,y). The image defogging effect of the present invention is good; the matrix addition operation is adopted, and the processing speed is fast.

Description

Technical field [0001] The present invention provides a method for defogging a single image, and specifically relates to a method for dehazing a single image based on deep learning. Background technique [0002] The impact of haze weather on drone aerial photography operations cannot be underestimated. The images formed by aerial photography in haze weather are blurred and the information that people need to capture is lost. As air conditions worsen, people's demand for dehazing drone aerial images is increasing. [0003] The problem of image defogging is an image restoration problem, which cannot be solved by simple image enhancement technology. According to McCartney's fogging model (E.J.McCartney, "Optics of the atmosphere:scattering by molecules and particles," New York, John Wiley and Sons, Inc., 1976.421p., 1976.), the scenery in the original scene is formed by passing through the air The water droplets or micro-particles are captured and enter the lens after refrac...

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 Patents(China)
IPC IPC(8): G06T5/00G06T5/50
CPCG06T5/50G06T5/73
Inventor 邹月娴陈泽晗王毅
Owner PEKING UNIV SHENZHEN GRADUATE SCHOOL