CN106023092A - Image defogging method and device - Google Patents
Image defogging method and device Download PDFInfo
- Publication number
- CN106023092A CN106023092A CN201610290289.4A CN201610290289A CN106023092A CN 106023092 A CN106023092 A CN 106023092A CN 201610290289 A CN201610290289 A CN 201610290289A CN 106023092 A CN106023092 A CN 106023092A
- Authority
- CN
- China
- Prior art keywords
- mist elimination
- image
- elimination image
- pixel
- treating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 239000011159 matrix material Substances 0.000 claims abstract description 75
- 230000008030 elimination Effects 0.000 claims description 178
- 238000003379 elimination reaction Methods 0.000 claims description 178
- 239000003595 mist Substances 0.000 claims description 165
- 238000002835 absorbance Methods 0.000 claims description 20
- 230000009466 transformation Effects 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 8
- 239000000203 mixture Substances 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 3
- 238000002834 transmittance Methods 0.000 abstract 2
- 238000005516 engineering process Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- GRYLNZFGIOXLOG-UHFFFAOYSA-N Nitric acid Chemical compound O[N+]([O-])=O GRYLNZFGIOXLOG-UHFFFAOYSA-N 0.000 description 1
- QAOWNCQODCNURD-UHFFFAOYSA-N Sulfuric acid Chemical compound OS(O)(=O)=O QAOWNCQODCNURD-UHFFFAOYSA-N 0.000 description 1
- 239000000443 aerosol Substances 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 150000002430 hydrocarbons Chemical class 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910017604 nitric acid Inorganic materials 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 235000011149 sulphuric acid Nutrition 0.000 description 1
- 239000001117 sulphuric acid Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention provides an image defogging method and device. The method comprises the steps that a to-be-defogged image is acquired; according to the to-be-defogged image, an atmospheric light value is acquired; the R channel matrix of the to-be-defogged image and the R channel frequency-domain matrix of the R channel matrix of the to-be-defogged image are acquired; the row position of each pixel in the R channel frequency-domain matrix and the column position of each pixel in the R channel frequency-domain matrix are acquired; according to the parameters, a weighted average filter formed by weighting a Gaussian filter, a Gaussian high-pass filter and a Gaussian band-stop filter is acquired; the R channel frequency-domain matrix and the weighted average filter are convoluted, and a convolution result is converted into a spatial domain to acquire a transmittance map; and the defogged image of the to-be-defogged image is acquired according to the atmospheric light value and the transmittance map. According to the invention, the R channel frequency-domain matrix and the weighted average filter are convoluted, so that the defogged image has high definition.
Description
Technical field
The present invention relates to technical field of image processing, particularly relate to a kind of image defogging method and device.
Background technology
Haze is a kind of natural phenomena, the small water droplet being suspended in a large number in surface air or the aerosol of ice crystal composition
System is referred to as mist, and dust in air, sulphuric acid, nitric acid, organic hydrocarbon compounds etc. can make the particle that air is muddy be referred to as haze.Mist
The existence of haze makes environment visibility reduce, and air quality deteriorates, and harm is serious；Also make picture imaging smudgy, to based on regarding
The self-navigation of frequency, monitor, the application of the technology such as tracking has certain inhibition.Therefore, image procossing and computer is utilized to regard
Feel technology, to image mist elimination, is allowed to sharpening, has very important significance.
Existing image defogging method includes two big classes: one is method based on image enhaucament, and this method does not consider figure
As the reason degenerated, but strengthen picture contrast, make visual effect relatively good；Two is image mist elimination based on physical model,
By setting up greasy weather degradation model, it is finally inversed by the image do not degenerated.Equations of The Second Kind method go fog effect more natural, image information
Amount loss is little, has the most associated research both at home and abroad.
Above-mentioned defogging method treatment effect has the phenomenon such as distortion, halation to occur, causes not fogging clear.
Summary of the invention
The present invention provides a kind of a kind of image mist elimination side overcoming the problems referred to above or solving the problems referred to above at least in part
Method and device.
First aspect, the present invention provides a kind of image defogging method, including:
Mist elimination image is treated in acquisition；
According to described mist elimination image for the treatment of, obtain air light value；
Treat the R access matrix of mist elimination image described in acquisition, and described in acquisition, treat the R passage of the R access matrix of mist elimination image
Frequency domain matrix；
Obtain in described R passage frequency domain matrix in the line position residing for each pixel and described R passage frequency domain matrix every
Column position residing for individual pixel；
According to the default number of lines of pixels treating mist elimination image, default described in treat that the pixel columns of mist elimination image, described R are logical
In road frequency domain matrix, column position residing for each pixel in line position residing for each pixel, described R passage frequency domain matrix, obtains
The weighted average filter formed is weighted by Gaussian filter, Gauss high pass filter and Gauss band elimination filter；
Described R passage frequency domain matrix and described weighted average filter are carried out convolution, and convolution results is converted into sky
Territory, to obtain absorbance figure；
According to described air light value and absorbance figure, described in acquisition, treat the mist elimination image of mist elimination image.
Preferably, obtain after treating mist elimination image, according to described mist elimination image for the treatment of, before obtaining air light value, described side
Method also includes:
Treat that mist elimination image is carried out sampling processing to described.
Preferably, according to described mist elimination image for the treatment of, obtain air light value, including:
The brightness value of the R of each pixel, G, channel B in mist elimination image is treated described in acquisition；
Obtain the minima in the brightness value of the R of described each pixel, G, channel B；
Determine the minima in the brightness value of the R of described each pixel, G, channel B corresponding passage composition image be institute
State the dark channel image treating mist elimination image；
According to the described dark channel image treating mist elimination image, obtain air light value.
Preferably, according to the described dark channel image treating mist elimination image, obtain air light value, including:
Obtain the brightness value of each pixel in described dark channel image, and by descending for the brightness value of described each pixel
Sequence；
In ranking results, choose the brightness value of above predetermined number, and according to the brightness value of described predetermined number, obtain
Air light value.
Preferably, according to the brightness value of described predetermined number, obtain air light value, including:
Obtain the brightness value of the R of each pixel, G, channel B in the first collection of pixels；Described first collection of pixels is described
In the brightness value of predetermined number corresponding to each brightness value described in treat the set that the pixel in mist elimination image is constituted；
Obtain the R of each pixel, G, the brightness value of channel B in described first collection of pixels；
The maximum brightness value determined in described first collection of pixels in the brightness value of the R of each pixel, G, channel B is air
Light value.
Preferably, described method also includes:
The described mist elimination image treating mist elimination image is carried out linear transformation, treats going of mist elimination image to obtain described in optimization
Mist image.
Preferably, according to the default number of lines of pixels treating mist elimination image, default described in treat mist elimination image pixel columns,
Row position residing for each pixel in line position residing for each pixel, described R passage frequency domain matrix in described R passage frequency domain matrix
Put, obtain and weighted the weighted average filter formed, bag by Gaussian filter, Gauss high pass filter and Gauss band elimination filter
Include:
According to the default number of lines of pixels treating mist elimination image, default described in treat that the pixel columns of mist elimination image, described R are logical
In road frequency domain matrix, column position residing for each pixel in line position residing for each pixel, described R passage frequency domain matrix, passes through
Formula (one) obtains and is weighted the weighted average filtering formed by Gaussian filter, Gauss high pass filter and Gauss band elimination filter
Device
H=w1H1+w2H2+w3H3 (one)
Wherein, H1 is Gaussian filter,K is constant, and σ is constant, and H2 is Gauss high pass filter,H3 is Gauss band elimination filter,Freq is constant,
Width is constant, and x is the line position in R passage frequency domain matrix residing for each pixel, and y is each pixel in R passage frequency domain matrix
Residing column position, u is the default number of lines of pixels treating mist elimination image, and v is the default pixel columns treating mist elimination image, w1、
w2, w be constant, and w1=w2=w3=1/3.
Preferably, according to described air light value and absorbance figure, described in acquisition, treat the mist elimination image of mist elimination image, including:
According to described air light value and absorbance figure, by treating the mist elimination image of mist elimination image described in formula (two) acquisition
Wherein, J (x) is the mist elimination image treating mist elimination image, and I (x) is for treating mist elimination image, and t (x) is absorbance figure, and A is big
Gas light value.
Preferably, the described mist elimination image treating mist elimination image is carried out linear transformation, treat mist elimination to obtain described in optimization
The mist elimination image of image, including:
By formula (three), the described mist elimination image treating mist elimination image is carried out linear transformation, treat described in optimization to obtain
The mist elimination image of mist elimination image
Idehaze=a Id+ b/255 (three)
Wherein, IdehazeFor the mist elimination image treating mist elimination image optimized, a be constant, b be constant, IdMist elimination is treated for described
The mist elimination image of image treats mist elimination image.
Second aspect, the invention provides a kind of image demister, including:
First acquiring unit, treats mist elimination image for acquisition；
Second acquisition unit, for treating mist elimination image described in basis, obtains air light value；
3rd acquiring unit, is used for treating described in obtaining the R access matrix of mist elimination image, and treats mist elimination image described in acquisition
The R passage frequency domain matrix of R access matrix；
4th acquiring unit, for obtaining in described R passage frequency domain matrix the line position residing for each pixel and described R
Column position residing for each pixel in passage frequency domain matrix；
5th acquiring unit, for according to preset the number of lines of pixels treating mist elimination image, default described in treat mist elimination image
Pixel columns, each picture in line position residing for each pixel, described R passage frequency domain matrix in described R passage frequency domain matrix
Column position residing for element, obtains and is weighted, by Gaussian filter, Gauss high pass filter and Gauss band elimination filter, the weighting formed
Average filter；
Convolution unit, for carrying out convolution by described R passage frequency domain matrix and described weighted average filter, and by convolution
Result is converted into spatial domain, to obtain absorbance figure；
6th acquiring unit, for according to described air light value and absorbance figure, treats the mist elimination of mist elimination image described in acquisition
Image.
As shown from the above technical solution, due to the fact that by by Gaussian filter, Gauss high pass filter and Gauss band
The weighted average filter of resistance wave filter composition carries out convolution with described R passage frequency domain matrix, thus makes the mist elimination image of acquisition
Definition is high and degree of refinement is more preferable, solves the problem such as image fault, halation in prior art.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to
Other accompanying drawing is obtained according to these figures.
The schematic flow sheet of a kind of image defogging method that Fig. 1 provides for one embodiment of the invention；
Fig. 2 is the structural representation of a kind of image demister that one embodiment of the invention provides.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments wholely.Based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under not making creative work premise
Embodiment, broadly falls into the scope of protection of the invention.
The schematic flow sheet of a kind of image defogging method that Fig. 1 provides for one embodiment of the invention.
As it is shown in figure 1, a kind of image defogging method of the present embodiment, including:
Mist elimination image is treated in S11, acquisition；
Treat mist elimination image described in S12, basis, obtain air light value；
Treat the R access matrix of mist elimination image described in S13, acquisition, and described in acquisition, treat the R of the R access matrix of mist elimination image
Passage frequency domain matrix；
It is understood that can by Fourier transformation the described R access matrix treating mist elimination image is converted to described in treat
The R passage frequency domain matrix of the R access matrix of mist elimination image.
S14, obtain in described R passage frequency domain matrix the line position residing for each pixel and described R passage frequency domain matrix
In column position residing for each pixel；
S15, according to preset the number of lines of pixels treating mist elimination image, default described in treat the pixel columns of mist elimination image, institute
State in R passage frequency domain matrix the row position residing for each pixel in the line position residing for each pixel, described R passage frequency domain matrix
Put, obtain and weighted, by Gaussian filter, Gauss high pass filter and Gauss band elimination filter, the weighted average filter formed；
S16, described R passage frequency domain matrix and described weighted average filter are carried out convolution, and convolution results is changed
Become spatial domain, to obtain absorbance figure；
It is understood that spatial domain can be converted into convolution results by Fourier inversion.
S17, according to described air light value and absorbance figure, treat the mist elimination image of mist elimination image described in acquisition.
Due to the fact that and put down by the weighting being made up of Gaussian filter, Gauss high pass filter and Gauss band elimination filter
All wave filter and described R passage frequency domain matrixes carry out convolution, thus make the mist elimination image definition of acquisition high and degree of refinement more
Good, solve the problem such as image fault, halation in prior art.
As a kind of preferred embodiment, described step S15, including:
According to the default number of lines of pixels treating mist elimination image, default described in treat that the pixel columns of mist elimination image, described R are logical
In road frequency domain matrix, column position residing for each pixel in line position residing for each pixel, described R passage frequency domain matrix, passes through
Formula (one) obtains and is weighted the weighted average filtering formed by Gaussian filter, Gauss high pass filter and Gauss band elimination filter
Device
H=w1H1+w2H2+w3H3 (one)
Wherein, H1 is Gaussian filter,K is constant, and σ is constant, and H2 is Gauss high pass filter,H3 is Gauss band elimination filter,Freq is constant,
Width is constant, and x is the line position in R passage frequency domain matrix residing for each pixel, and y is each pixel in R passage frequency domain matrix
Residing column position, u is the default number of lines of pixels treating mist elimination image, and v is the default pixel columns treating mist elimination image, w1、
w2, w be constant, and w1=w2=w3=1/3.
In actual applications, usually, k=1.2, σ=50, freq=100, width=10,
As a kind of preferred embodiment, described step S17, including:
According to described air light value and absorbance figure, by treating the mist elimination image of mist elimination image described in formula (two) acquisition
Wherein, J (x) is the mist elimination image treating mist elimination image, and I (x) is for treating mist elimination image, and t (x) is absorbance figure, and A is big
Gas light value.
As a kind of preferred embodiment, after described step S11, before described step S12, described method also includes:
Treat that mist elimination image is carried out sampling processing to described.
In the present embodiment, can treat that mist elimination image carries out interlacing and takes pixel value every row to described, obtain artwork 1/4 size
Down-sampled image.
Owing to reducing the pixel count treating mist elimination image, treatment effeciency therefore can be improved.
What deserves to be explained is, this kind of process does not interferes with the definition of mist elimination image.
As a kind of preferred embodiment, described step S12, including:
The brightness value of the R of each pixel, G, channel B in mist elimination image is treated described in acquisition；
Obtain the minima in the brightness value of the R of described each pixel, G, channel B；
Determine the minima in the brightness value of the R of described each pixel, G, channel B corresponding passage composition image be institute
State the dark channel image treating mist elimination image；
According to the described dark channel image treating mist elimination image, obtain air light value.
As a kind of preferred embodiment, in described step S12 according to described in treat the dark channel image of mist elimination image, obtain
Air light value, including:
Obtain the brightness value of each pixel in described dark channel image, and by descending for the brightness value of described each pixel
Sequence；
In ranking results, choose the brightness value of above predetermined number, and according to the brightness value of described predetermined number, obtain
Air light value.
As a kind of preferred embodiment, the brightness value according to described predetermined number in described step S12, obtain atmosphere light
Value, including:
Obtain the brightness value of the R of each pixel, G, channel B in the first collection of pixels；Described first collection of pixels is described
In the brightness value of predetermined number corresponding to each brightness value described in treat the set that the pixel in mist elimination image is constituted；
Obtain the maximum brightness value in the brightness value of the R of each pixel, G, channel B in described first collection of pixels；
The maximum brightness value determined in described first collection of pixels in the brightness value of the R of each pixel, G, channel B is air
Light value.
As a kind of preferred embodiment, described method also includes:
The described mist elimination image treating mist elimination image is carried out linear transformation, treats going of mist elimination image to obtain described in optimization
Mist image.
In the present embodiment, by formula (three), the described mist elimination image treating mist elimination image is carried out linear transformation, to obtain
Take the mist elimination image treating mist elimination image described in optimization
Idehaze=a Id+ b/255 (three)
Wherein, IdehazeFor the mist elimination image treating mist elimination image optimized, a be constant, b be constant, IdMist elimination is treated for described
The mist elimination image of image treats mist elimination image.
In actual applications, usually, a=1.4, b=-20.
It is understood that can improve the contrast of mist elimination image through linear transformation, so that mist elimination image is more clear
Clear.
Fig. 2 is the structural representation of a kind of image demister that one embodiment of the invention provides.
As in figure 2 it is shown, a kind of image demister, including: the first acquiring unit 21, second acquisition unit the 22, the 3rd obtain
Take unit the 23, the 4th acquiring unit the 24, the 5th acquiring unit 25, convolution unit 26 and the 7th acquiring unit 27, wherein,
First acquiring unit 21 treats mist elimination image for acquisition；
Second acquisition unit 22, for treating mist elimination image described in basis, obtains air light value；
3rd acquiring unit 23 treats the R access matrix of mist elimination image described in obtain, and treats mist elimination image described in obtaining
The R passage frequency domain matrix of R access matrix；
4th acquiring unit 24 is for obtaining in described R passage frequency domain matrix the line position residing for each pixel and described
Column position residing for each pixel in R passage frequency domain matrix；
5th acquiring unit 25 for according to the number of lines of pixels treating mist elimination image preset, default described in treat mist elimination image
Pixel columns, each picture in line position residing for each pixel, described R passage frequency domain matrix in described R passage frequency domain matrix
Column position residing for element, obtains and is weighted, by Gaussian filter, Gauss high pass filter and Gauss band elimination filter, the weighting formed
Average filter；
Convolution unit is used for 26 and described R passage frequency domain matrix and described weighted average filter carries out convolution, and will volume
Long-pending result is converted into spatial domain, to obtain absorbance figure；
6th acquiring unit 27, for according to described air light value and absorbance figure, treats the mist elimination of mist elimination image described in acquisition
Image.
A kind of image demister of the present invention and a kind of image defogging method one_to_one corresponding, be not described in detail in this.
It should be noted that, in all parts of assembly of the invention, the function to be realized according to it and to therein
Parts have carried out logical partitioning, but, the present invention is not only restricted to this, can as required all parts be repartitioned or
Person combines, for example, it is possible to be single parts by some unit constructions, or can be further broken into more by some parts
Subassembly.
The all parts embodiment of the present invention can realize with hardware, or to run on one or more processor
Software module realize, or with combinations thereof realize.It will be understood by those of skill in the art that and can use in practice
Microprocessor or digital signal processor (DSP) realize the some or all portions in device according to embodiments of the present invention
The some or all functions of part.The present invention is also implemented as the part for performing method as described herein or complete
The equipment in portion or device program (such as, computer program and computer program).Such program realizing the present invention
Can store on a computer-readable medium, or can be to have the form of one or more signal.Such signal is permissible
Download from internet website and obtain, or provide on carrier signal, or provide with any other form.
The present invention will be described rather than limits the invention to it should be noted above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.Word " includes " not arranging
Except there is the element or step not arranged in the claims.The present invention can by means of include the hardware of some different elements with
And realize by means of properly programmed computer.If in the unit claim listing equipment for drying, in these devices
Several can be specifically to be embodied by same hardware branch.
Embodiment of above is only suitable to illustrate the present invention, and not limitation of the present invention, common about technical field
Technical staff, without departing from the spirit and scope of the present invention, it is also possible to make a variety of changes and modification, therefore own
The technical scheme of equivalent falls within scope of the invention, and the scope of patent protection of the present invention should be defined by the claims.
Claims (10)
1. an image defogging method, it is characterised in that including:
Mist elimination image is treated in acquisition；
According to described mist elimination image for the treatment of, obtain air light value；
Treat the R access matrix of mist elimination image described in acquisition, and described in acquisition, treat the R passage frequency domain of the R access matrix of mist elimination image
Matrix；
Obtain in described R passage frequency domain matrix each picture in the line position residing for each pixel and described R passage frequency domain matrix
Column position residing for element；
According to the default number of lines of pixels treating mist elimination image, default described in treat the pixel columns of mist elimination image, described R passage frequency
In domain matrix, column position residing for each pixel in line position residing for each pixel, described R passage frequency domain matrix, obtains by height
The weighted average filter that this wave filter, Gauss high pass filter and the weighting of Gauss band elimination filter are formed；
Described R passage frequency domain matrix and described weighted average filter are carried out convolution, and convolution results is converted into spatial domain, with
Obtain absorbance figure；
According to described air light value and absorbance figure, described in acquisition, treat the mist elimination image of mist elimination image.
Method the most according to claim 1, it is characterised in that obtain after treating mist elimination image, treat mist elimination figure according to described
Picture, before obtaining air light value, described method also includes:
Treat that mist elimination image is carried out sampling processing to described.
Method the most according to claim 1 and 2, it is characterised in that according to described mist elimination image for the treatment of, obtains air light value,
Including:
The brightness value of the R of each pixel, G, channel B in mist elimination image is treated described in acquisition；
Obtain the minima in the brightness value of the R of described each pixel, G, channel B；
Determine the minima in the brightness value of the R of described each pixel, G, channel B corresponding passage composition image be described in treat
The dark channel image of mist elimination image；
According to the described dark channel image treating mist elimination image, obtain air light value.
Method the most according to claim 3, it is characterised in that according to the described dark channel image treating mist elimination image, obtains
Air light value, including:
Obtain the brightness value of each pixel in described dark channel image, and by the descending row of brightness value of described each pixel
Sequence；
In ranking results, choose the brightness value of above predetermined number, and according to the brightness value of described predetermined number, obtain air
Light value.
Method the most according to claim 4, it is characterised in that according to the brightness value of described predetermined number, obtains atmosphere light
Value, including:
Obtain the brightness value of the R of each pixel, G, channel B in the first collection of pixels；Described first collection of pixels is described presetting
In the brightness value of quantity corresponding to each brightness value described in treat the set that the pixel in mist elimination image is constituted；
Obtain the R of each pixel, G, the brightness value of channel B in described first collection of pixels；
The maximum brightness value determined in described first collection of pixels in the brightness value of the R of each pixel, G, channel B is atmosphere light
Value.
Method the most according to claim 1 and 2, it is characterised in that described method also includes:
The described mist elimination image treating mist elimination image is carried out linear transformation, to obtain the mist elimination figure treating mist elimination image described in optimization
Picture.
Method the most according to claim 1 and 2, it is characterised in that according to the default number of lines of pixels treating mist elimination image, in advance
Treat described in if in the pixel columns of mist elimination image, described R passage frequency domain matrix that the line position residing for each pixel, described R lead to
Column position residing for each pixel in road frequency domain matrix, obtains and is hindered filter by Gaussian filter, Gauss high pass filter and Gauss band
The weighted average filter that the weighting of ripple device is formed, including:
According to the default number of lines of pixels treating mist elimination image, default described in treat the pixel columns of mist elimination image, described R passage frequency
In domain matrix, column position residing for each pixel in line position residing for each pixel, described R passage frequency domain matrix, passes through formula
(1) weighted average filter formed by Gaussian filter, Gauss high pass filter and the weighting of Gauss band elimination filter is obtained
H=w1H1+w2H2+w3H3 (one)
Wherein, H1 is Gaussian filter,K is constant, and σ is constant, and H2 is Gauss high pass filter,H3 is Gauss band elimination filter,Freq is constant,
Width is constant, and x is the line position in R passage frequency domain matrix residing for each pixel, and y is each pixel in R passage frequency domain matrix
Residing column position, u is the default number of lines of pixels treating mist elimination image, and v is the default pixel columns treating mist elimination image, w1、
w2, w be constant, and w1=w2=w3=1/3.
Method the most according to claim 1 and 2, it is characterised in that according to described air light value and absorbance figure, obtains institute
State the mist elimination image treating mist elimination image, including:
According to described air light value and absorbance figure, by treating the mist elimination image of mist elimination image described in formula (two) acquisition
Wherein, J (x) is the mist elimination image treating mist elimination image, and I (x) is for treating mist elimination image, and t (x) is absorbance figure, and A is atmosphere light
Value.
Method the most according to claim 6, it is characterised in that the described mist elimination image treating mist elimination image is linearly become
Change, to obtain the mist elimination image treating mist elimination image described in optimization, including:
By formula (three), the described mist elimination image treating mist elimination image is carried out linear transformation, treat mist elimination to obtain described in optimization
The mist elimination image of image
Idehaze=a Id+ b/255 (three)
Wherein, IdehazeFor the mist elimination image treating mist elimination image optimized, a be constant, b be constant, IdMist elimination image is treated for described
Mist elimination image treat mist elimination image.
10. an image demister, it is characterised in that including:
First acquiring unit, treats mist elimination image for acquisition；
Second acquisition unit, for treating mist elimination image described in basis, obtains air light value；
3rd acquiring unit, is used for treating described in obtaining the R access matrix of mist elimination image, and treats described in acquisition that the R of mist elimination image leads to
The R passage frequency domain matrix of road matrix；
4th acquiring unit, for obtaining in described R passage frequency domain matrix the line position residing for each pixel and described R passage
Column position residing for each pixel in frequency domain matrix；
5th acquiring unit, for according to the number of lines of pixels treating mist elimination image preset, default described in treat the picture of mist elimination image
Each pixel institute in line position residing for each pixel, described R passage frequency domain matrix in element columns, described R passage frequency domain matrix
The column position at place, obtains and is weighted, by Gaussian filter, Gauss high pass filter and Gauss band elimination filter, the weighted average formed
Wave filter；
Convolution unit, for carrying out convolution by described R passage frequency domain matrix and described weighted average filter, and by convolution results
It is converted into spatial domain, to obtain absorbance figure；
6th acquiring unit, for according to described air light value and absorbance figure, treats the mist elimination image of mist elimination image described in acquisition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610290289.4A CN106023092B (en) | 2016-05-04 | 2016-05-04 | Image defogging method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610290289.4A CN106023092B (en) | 2016-05-04 | 2016-05-04 | Image defogging method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106023092A true CN106023092A (en) | 2016-10-12 |
| CN106023092B CN106023092B (en) | 2020-12-11 |
Family
ID=57082073
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610290289.4A Active CN106023092B (en) | 2016-05-04 | 2016-05-04 | Image defogging method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106023092B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111127340A (en) * | 2019-12-05 | 2020-05-08 | 南京工程学院 | Image defogging method |
| CN111383181A (en) * | 2018-12-28 | 2020-07-07 | 展讯通信（上海）有限公司 | Image enhancement method and device, storage medium and terminal |
| WO2021046743A1 (en) * | 2019-09-11 | 2021-03-18 | Covidien Lp | Systems and methods for smoke-reduction in images |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013029337A1 (en) * | 2011-08-30 | 2013-03-07 | Fujitsu Limited | Image defogging method and system |
| CN102968772A (en) * | 2012-12-04 | 2013-03-13 | 电子科技大学 | Image defogging method based on dark channel information |
| CN103034983A (en) * | 2013-01-10 | 2013-04-10 | 厦门大学 | Defogging method based on anisotropic filtering |
| CN103177424A (en) * | 2012-12-07 | 2013-06-26 | 西安电子科技大学 | Low-luminance image reinforcing and denoising method |
| CN104036466A (en) * | 2014-06-17 | 2014-09-10 | 浙江立元通信技术股份有限公司 | Video defogging method and system |
| CN104091307A (en) * | 2014-06-12 | 2014-10-08 | 中国人民解放军重庆通信学院 | Frog day image rapid restoration method based on feedback mean value filtering |
| CN105023256A (en) * | 2015-08-13 | 2015-11-04 | 丘璇 | Image defogging method and system |
-
2016
- 2016-05-04 CN CN201610290289.4A patent/CN106023092B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013029337A1 (en) * | 2011-08-30 | 2013-03-07 | Fujitsu Limited | Image defogging method and system |
| CN102968772A (en) * | 2012-12-04 | 2013-03-13 | 电子科技大学 | Image defogging method based on dark channel information |
| CN103177424A (en) * | 2012-12-07 | 2013-06-26 | 西安电子科技大学 | Low-luminance image reinforcing and denoising method |
| CN103034983A (en) * | 2013-01-10 | 2013-04-10 | 厦门大学 | Defogging method based on anisotropic filtering |
| CN104091307A (en) * | 2014-06-12 | 2014-10-08 | 中国人民解放军重庆通信学院 | Frog day image rapid restoration method based on feedback mean value filtering |
| CN104036466A (en) * | 2014-06-17 | 2014-09-10 | 浙江立元通信技术股份有限公司 | Video defogging method and system |
| CN105023256A (en) * | 2015-08-13 | 2015-11-04 | 丘璇 | Image defogging method and system |
Non-Patent Citations (1)
| Title |
|---|
| 刘丰: "图像去雾算法研究及硬件实现", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111383181A (en) * | 2018-12-28 | 2020-07-07 | 展讯通信（上海）有限公司 | Image enhancement method and device, storage medium and terminal |
| CN111383181B (en) * | 2018-12-28 | 2022-09-27 | 展讯通信（上海）有限公司 | Image enhancement method and device, storage medium and terminal |
| WO2021046743A1 (en) * | 2019-09-11 | 2021-03-18 | Covidien Lp | Systems and methods for smoke-reduction in images |
| CN111127340A (en) * | 2019-12-05 | 2020-05-08 | 南京工程学院 | Image defogging method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106023092B (en) | 2020-12-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103218622B (en) | A kind of haze monitoring method based on computer vision | |
| Zhang et al. | Single image defogging based on multi-channel convolutional MSRCR | |
| CN106023092A (en) | Image defogging method and device | |
| CN103955905B (en) | Based on the single image to the fog method that fast wavelet transform and weighted image merge | |
| Su et al. | Transformer-based regression network for pansharpening remote sensing images | |
| CN101103378B (en) | Device and method for creating a saliency map of an image | |
| CN105096280A (en) | Method and device for processing image noise | |
| CN104616258B (en) | A kind of rapid defogging method for road image | |
| CN106778705A (en) | A kind of pedestrian's individuality dividing method and device | |
| WO2006086102A1 (en) | System and method for detecting features from images of vehicles | |
| CN105741296B (en) | 360 degree of full views of one kind get a bird's eye view panorama driving auxiliary calibration method | |
| CN105354806A (en) | Dark channel based rapid defogging method and system | |
| CN108022222A (en) | A kind of thin cloud in remote sensing image minimizing technology based on convolution-deconvolution network | |
| Wen et al. | An effective network integrating residual learning and channel attention mechanism for thin cloud removal | |
| CN106251296A (en) | A kind of image defogging method and system | |
| Zhu et al. | Infrared moving point target detection based on an anisotropic spatial-temporal fourth-order diffusion filter | |
| CN109801231A (en) | Image processing method of electrophoresis electronic paper detection equipment | |
| CN112651469A (en) | Infrared and visible light image fusion method and system | |
| CN109064402B (en) | Single image super-resolution reconstruction method based on enhanced non-local total variation model prior | |
| CN110033439A (en) | The visible detection method of belt conveyer material blocking in a kind of Primary Processing | |
| CN106971377A (en) | A kind of removing rain based on single image method decomposed based on sparse and low-rank matrix | |
| CN109934781A (en) | Image processing method, device, terminal device and computer readable storage medium | |
| CN109709597A (en) | The gain correction approach of flat panel detector | |
| CN106846250A (en) | A kind of super resolution ratio reconstruction method based on multi-scale filtering | |
| CN110211082A (en) | A kind of image interfusion method, device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |