Defogging fusion processing method based on polarization image data

A polarization image fusion processing technology, applied in image data processing, image enhancement, climate sustainability, etc., can solve the problems of inconsistency between the specific process of atmospheric transmission and model parameters, weak robustness, lack of atmospheric particle scattering information, etc. , to achieve good haze removal performance, enhance picture quality, and improve image quality

Active Publication Date: 2022-08-02
CHINA UNIV OF MINING & TECH (BEIJING)
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The traditional method is mainly based on light intensity images. The traditional method lacks sufficient atmospheric particle scattering information. In the low-contrast imaging environment after haze particle scattering, it is urgent to introduce information other than intensity and spectrum to restore clear images.
At the same time, for the situation where the haze environment is quite different, the method based on the atmospheric physical scattering model has inconsistent atmospheric transmission process and model parameters, and the robustness is not strong. Therefore, an effective and stable haze removal method is urgently needed to restore clarity. image

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
  • Defogging fusion processing method based on polarization image data
  • Defogging fusion processing method based on polarization image data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] A dehazing fusion processing method based on polarized image data, the method is as follows:

[0038] A. Polarization haze image collection: use a polarization camera to collect four haze polarization images in the same area (the preferred polarization camera of the present invention is formed by a combination of a color industrial camera and a polarization sensor, and the present invention uses four haze images of the same area Polarize the image and obtain the dehazed image L(x) according to the method, and the same area can also be dehazed and fused), the haze polarized image is an RGB color image, and the polarization angles of the four haze polarized images are 0°, 45°, 90°, 135° and expressed as I0, I45, I90, I145 in turn, the four haze polarized images are packaged into a set of polarized image packages, and the polarized image package is obtained according to the Stokes parameter according to the following formula The light intensity image of:

[0039]S 0 =(I0...

Embodiment 2

[0047] A dehazing fusion processing method based on polarized image data, the method is as follows:

[0048] A. Polarization haze image acquisition: Use a polarization camera to collect four haze polarization images in the same area. The haze polarization images are RGB color images. The polarization angles of the four haze polarization images are 0°, 45°, and 90° in turn. , 135° and expressed as I0, I45, I90, I145 in turn, package the four haze polarized images into a set of polarized image packages, and obtain the light intensity of the polarized image package according to the following formula according to the Stokes parameter (that is, the stocks parameter). image:

[0049] S 0 =(I0+I45+I90+I145) / 2;

[0050] In polarized light detection, the V component in the stocks component is considered to be 0 in general research, detection and calculation because of its small value. In this case, the observed light is considered to be a combination of linearly polarized light and ...

Embodiment 3

[0072] A dehazing fusion processing method based on polarized image data, the method is as follows:

[0073] A. Polarization haze image acquisition: use a polarization camera to collect four haze polarization images in the same area (the polarization camera is composed of a color industrial camera and a polarization sensor), the haze polarization image is an RGB color image, and the four haze polarization images are The polarization angles of the images are 0°, 45°, 90°, and 135° in turn, and they are expressed as I0, I45, I90, and I145. The four haze polarization images are packaged into a set of polarization image packages. According to the Stokes parameter ( That is, stocks parameter) to obtain the light intensity image of the polarization image package according to the following formula:

[0074] S 0 =(I0+I45+I90+I145) / 2;

[0075] B. Build a dehazing network model. The dehazing network model includes J-Net sub-network, T-Net sub-network and A-Net sub-network. The dehazin...

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 defogging fusion processing method based on polarization image data, and the method comprises the following steps: A, collecting four haze polarization images in the same region through a polarization camera, and packaging the four haze polarization images into a polarization image package; B, constructing a defogging network model comprising a J-Net sub-network, a T-Net sub-network and an A-Net sub-network, and processing the polarization image package; and C, the defogging network model outputs a defogged image L (x) according to the following formula: a polarization camera is adopted to collect four haze polarization images in the same area to form a polarization image packet, the defogging network model is constructed and trained to achieve the minimum image quality loss, and the defogged image is output based on an atmospheric scattering physical model. The visibility and contrast of the haze image are effectively improved, image information lost due to haze can be recovered, and the image quality is improved.

Description

technical field [0001] The invention relates to the fields of polarization optics and haze image processing, in particular to a dehazing fusion processing method based on polarization image data. Background technique [0002] As the visual basis for human perception of the world, images are an important way for humans to obtain information, express information and transmit information. However, haze particles have strong scattering and strong absorption effects on light, which will lead to problems such as chromatic aberration, image degradation, and reduced visibility in the collected images, which cause great trouble to people's daily life and health. Haze image restoration is the basis of image visual analysis and is crucial for obtaining image information. Traditional methods are mainly based on light intensity images. Traditional methods lack sufficient atmospheric particle scattering information. In low-contrast imaging environments after haze particle scattering, it ...

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): G06T5/00G06T5/50
CPCG06T5/50G06T2207/10004G06T5/73Y02A90/10
Inventor 陈伟毕朋帅杨澜
Owner CHINA UNIV OF MINING & TECH (BEIJING)
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