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

A method of image defogging

An image and input image technology, applied in the field of image defogging, can solve problems such as error, inaccurate estimation, loss of image information, etc., and achieve the effect of avoiding light spots

Active Publication Date: 2022-05-27
SHANGHAI DIANJI UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But there will be white fog blocks for the depth of field area; literature [7] In order to overcome the image block effect of dehazing images, a non-local prior algorithm is proposed and the color distance is captured frame by frame, but the assumption of no haze in the future may have errors in practice; literature [8] A defogging algorithm with adaptive transmittance is proposed to solve the problem of inaccurate estimation of the traditional dark channel prior transmittance, but there is a problem of image information loss when using the average gradient value to judge the fog concentration

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 method of image defogging
  • A method of image defogging
  • A method of image defogging

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0054] Example: as figure 1 As shown, the embodiment of the present invention includes an image dehazing method for processing a foggy image, and the physical model adopted by the method is a mathematical model of the fogged image, which is defined as the following formula:

[0055] I(x)=J(x)t(x)+A(1-t(x)) (1)

[0056] In the formula, I is the original foggy image captured, J is the restored clear fog-free image, t is the transmittance of light, A is the atmospheric light value, and x is the index of the degree of fogging.

[0057] In most non-sky local areas, some pixels will always have values ​​with very low (close to 0) color channels. Therefore, for any image J the dark channel can be defined as:

[0058]

[0059] In the formula, J c is a color channel of J, Ω(x) is the area centered on pixel x, J dark It is the dark primary color of J.

[0060] According to formula (1) and the dark primary color a priori, the minimum value of formula (1) is calculated twice to ob...

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. The method comprises: acquiring atmospheric light value from a foggy image; acquiring a transmittance image from the foggy image by using a guided filtering algorithm based on a canny operator; according to the atmospheric light value The fogged image is processed pixel by pixel with the transmittance image to obtain the dehazed image. This method can solve the local halo effect. This algorithm is superior to other algorithms in subjective visual effect, and the image edge denoising is more natural; the comparison of objective performance indicators shows that the improved dehazing algorithm has higher image quality and better image quality. Resilience.

Description

technical field [0001] The invention relates to the field of digital image processing, in particular to an image dehazing method. Background technique [0002] In haze weather or low illumination, the image recognition effect will be greatly affected. Therefore, the study of dehazing algorithm can improve the recognition efficiency. At present, there are mainly two types of dehazing methods. One is a dehazing algorithm based on a non-physical model, which uses an image enhancement algorithm to improve the image contrast to achieve a dehazing effect. The common one is based on histogram equalization. [1] and an algorithm based on the Retinex principle [2,3] , but this method will lose image information; the other is the dehazing algorithm based on physical model, mainly the classic dark primary color prior dehazing algorithm proposed by Dr. He Kaiming [4] , to restore the image, this method has a good dehazing effect, but there is also the problem of using the Soft Matting...

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): G06T5/00G06T7/155
CPCG06T7/155G06T5/73
Inventor 林锦晨杨俊杰于婷耿传平
Owner SHANGHAI DIANJI UNIV