CN107248146A - A kind of UUV Layer Near The Sea Surfaces visible images defogging method - Google Patents
A kind of UUV Layer Near The Sea Surfaces visible images defogging method Download PDFInfo
- Publication number
- CN107248146A CN107248146A CN201710362665.0A CN201710362665A CN107248146A CN 107248146 A CN107248146 A CN 107248146A CN 201710362665 A CN201710362665 A CN 201710362665A CN 107248146 A CN107248146 A CN 107248146A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- image
- matrix
- value
- 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 27
- 239000011159 matrix material Substances 0.000 claims abstract description 22
- 230000003044 adaptive effect Effects 0.000 claims abstract description 12
- 238000003384 imaging method Methods 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000007935 neutral effect Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 abstract description 7
- 230000006870 function Effects 0.000 description 10
- 230000000694 effects Effects 0.000 description 4
- 206010027146 Melanoderma Diseases 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000003595 mist Substances 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000009738 saturating Methods 0.000 description 1
- 239000013535 sea water Substances 0.000 description 1
- 230000007704 transition 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The present invention is to provide a kind of UUV Layer Near The Sea Surfaces visible images defogging method.(1) original foggy image is obtained, the minimum value of three Color Channels is asked for, is designated as M (x);(2) global atmosphere light is asked for using M (x) and quaternary tree close classification improved method;(3) adaptive surface is carried out to M (x) to obscure;(4) atmospheric scattering function is obtained;(5) fog free images are tried to achieve by greasy weather Imaging physics model and exported.The present invention carries out adaptive surface to the minimum value image of three Color Channels and obscured, and each pixel has the convolution matrix of oneself, can preferably keep the edge of target object, is conducive to follow-up target identification, positioning and tracks.Meanwhile, in order to overcome local dark phenomenon partially after filtering, image is roughly divided into dark areas and bright area by global atmosphere light, Fuzzy Processings are carried out using different radiuses respectively.The invention defogging degree is preferable and can overcome sky halation phenomenon.
Description
Technical field
The present invention relates to a kind of smooth visual imaging method, especially a kind of defogging of UUV Layer Near The Sea Surfaces visible images
Method.
Background technology
China sea water front is very long, and marine weather conditions are complicated, because ship collision accident is of common occurrence caused by sea fog.Light
Particle refraction is suspended when being propagated in atmospheric medium or is reflected, causes the video camera being arranged on maritime affairs buoy or UUV to be clapped
Big measure feature in the video image taken the photograph is blanked, and contrast declines, and edge details are obscured, and easily cause the mistake of picture material
Sentence, so as to be impacted to follow-up target positioning, recognition and tracking.Therefore, the defogging processing of marine light visual pattern is one
The problem of item is practical and challenging.
Document《Single Image Haze Removal Using Dark Channel Prior》(Proceedings
Of IEEE CVPR.Miami, USA:IEEE Computer Society,2009:One kind is proposed in 2341-2353) to be based on
The defogging method of dark channel prior.Go out transmissivity first with mini-value filtering rough estimate, then using soft pick figure algorithm to saturating
The rate of penetrating is refined, and then recovers fog free images.But, statistics rule of the dark primary priori based on fog free images, if mesh
Mark in scene in just similar with atmosphere light, such as sea white wave, sky areas or bright areas, due to precondition not just
Really, cause the transmissivity estimated in these regions too small, cause the fog free images large area distortion after recovering or local appearance black
Spot phenomenon.Therefore, the method based on dark defogging can not adapt to water surface Misty Image completely.
Document《Fast Visibility Restoration from a Single Color or Gray Level
Image》(Proceedings of IEEE ICCV,2009:2201-2208) the deformation type estimation air of medium filtering is utilized in
Dissipative function, it is assumed that atmospheric dissipation function approaches maximum in feasible zone, and localized variation is gentle, proposes a kind of rapid image
Defogging method.But, the not good holding edge filter method of medium filtering.It is another to have much based on dark and greasy weather imaging thing
The improved method for managing model.
The content of the invention
Image border can preferably be retained it is an object of the invention to provide one kind, defogging degree is good and can overcome
The UUV Layer Near The Sea Surface visible images defogging methods of sky halation phenomenon.
The object of the present invention is achieved like this:
(1) original foggy image is obtained, the minimum value of three Color Channels is asked for, is designated as M (x);
(2) global atmosphere light is asked for using M (x) and quaternary tree close classification improved method;
(3) adaptive surface is carried out to M (x) to obscure;
(4) atmospheric scattering function is obtained;
(5) fog free images are tried to achieve by greasy weather Imaging physics model and exported.
The present invention can also include:
1st, the improved method of the utilization M (x) and quaternary tree close classification obtains global atmosphere light and specifically included:
1) image is divided into four rectangular areas;
2) average pixel value and variance in each region are asked for, average pixel value subtracts variance and is considered as each area score;
3) it is four zonules by the region division of highest scoring;
4) repeat step 2)~3), until the area in top score region is less than threshold value S set in advanceT;
5) the most bright spot in the region is taken, its pixel value is global atmosphere light A.
2nd, it is described to specifically including that M (x) progress adaptive surfaces are obscured:
Adaptive surface Fuzzy Processing has two parameters, radius s and threshold value T, and radius s determines fuzzy scope, and threshold value T determines
The degree of cover half paste, fuzzy scope is exactly the size of convolution matrix, when radius is s, a diameter of 2s+1 of fuzzy matrix, matrix
Element number is s*s, and the neutral element of matrix is current pixel value, and the calculation formula of matrix element value is:
Wherein T is threshold value, wijFor the element value of pattern matrix, be also weight, MijFor image M (x) pixel values, M0For template
The pixel value at matrix center;
Then to wijPre-process
wij=max (1, wij)
According to convolution algorithm, each pixel is by the value after obscuring:
As the pixel value M at pattern matrix center0It is bright areas, blur radius s=3, window size is 7* during > A-50
7;As the pixel value M at pattern matrix center0During≤A-50, for local dark areas, windows radius s=1, using 3*3 window.
The beneficial effects of the invention are as follows:The minimum value image progress adaptive surface of three Color Channels is obscured.It with
Other process of convolution means such as medium filtering, mean filter are different, and each pixel of this method has the convolution matrix of oneself, energy
The edge of enough preferably holding target objects, is conducive to follow-up target identification, positioning and tracks.Meanwhile, in order to overcome filtering
Local dark phenomenon partially, is roughly divided into dark areas and bright area, respectively using different by global atmosphere light by image afterwards
Radius carries out Fuzzy Processing.The invention defogging degree is preferable and can overcome sky halation phenomenon.
Brief description of the drawings
Fig. 1 is flow chart of the invention;
Fig. 2 is the primary visible light foggy image of one embodiment of the invention;
Fig. 3 is that one embodiment of the invention takes the image M (x) after three Color Channel minimum values;
When Fig. 4 is that one embodiment of the invention obtains global atmosphere light using the improved method of M (x) and quaternary tree close classification
Region segmentation schematic diagram;
Fig. 5 is images of the M (x) of one embodiment of the invention through adaptive surface after fuzzy;
Fig. 6 is the image of the atmospheric scattering function of one embodiment of the invention;
Fig. 7 is the mist elimination image of one embodiment of the invention.
Embodiment
In order that the object, technical solutions and advantages of the present invention become apparent from understanding, below in conjunction with accompanying drawing and specific implementation
Example, is described in further detail to the present invention.
As shown in figure 1, a kind of UUV Layer Near The Sea Surfaces visible images defogging method, comprises the following steps:
1. being inputted using VS2010 and OPENCV Programming with Pascal Language and showing that UUV Layer Near The Sea Surfaces visible ray investigates image, I is designated as
(x), it is seen that light image is as shown in Figure 2.The minimum value of three Color Channels is taken to I (x), is remembered
Image M (x) is as shown in Figure 3.
2. the step of obtaining global atmosphere light using the improved method of M (x) and quaternary tree close classification includes:
(1) picture is divided into four rectangular areas.
(2) the pixel average value and variance in each region are asked for, average pixel value subtracts variance and is considered as each area score.
(3) it is four zonules by the region division of highest scoring.
(4) repeat step (2)~(3), until the area in top score region is less than threshold value S set in advanceT.Through experiment
Know, 320*240 test pictures, STDuring=5*3.75, the coordinate of most bright spot just can be tried to achieve in the case where iterations is minimum.
Region segmentation schematic diagram is as shown in Figure 4.
(5) the most bright spot in the region is taken, its pixel value is global atmosphere light A.
3. couple M (x), which carries out the fuzzy specific steps of adaptive surface, to be included:
The fuzzy process of convolution means such as from medium filtering, mean filter of adaptive surface are different, each pixel of this method
There is the convolution matrix of oneself.It mainly has two parameters, radius s and threshold value T, and the former determines fuzzy scope, and the latter determines
Fuzzy degree.Fuzzy scope is exactly the size of convolution matrix, when radius is s, a diameter of 2s+1 of fuzzy matrix, matrix element
Plain number is s*s, and the neutral element of matrix is current pixel value.The calculation formula of matrix element value is:
Wherein T is threshold value, wijIt is also weight, M for the element value of pattern matrixijFor image M (x) pixel values, M0For template
The pixel value at matrix center.Then to wijDo a pretreatment
wij=max (1, wij) (3)
According to convolution algorithm, each pixel is by the value after obscuring:
If entire image carries out Fuzzy Processing using with Radius, it is found that after defogging at image border, details is excessively
It is prominent, there is transition between " burr " phenomenon, pixel unnatural, and filtered image some regions can produce black patch.This is
Because at the edge of Strength Changes, if certain pixel intensity level is low, but most pixel intensity values are larger in window, after obscuring
The point takes larger intensity level.Therefore, for the relatively low local dark areas of intensity level, prevent the intensity level because window in other
Pixel is excessively amplified, and former filter radius is reduced.Therefore, roughly entire image is divided into by global atmosphere light A
Bright area and dark areas.As the pixel value M at pattern matrix center0It is bright areas, blur radius s=3, window during > A-50
Size is 7*7;As the pixel value M at pattern matrix center0During≤A-50, for local dark areas, windows radius s=1, using 3*
3 window, reduces the phenomenon that border area pixels are excessively amplified, and removes blackspot effect.After M (x) is obscured through adaptive surface
Image Mblur(x) as shown in Figure 5.
4. obtaining the specific steps of atmospheric scattering function includes:
First, the average ave of M (x) all elements is sought.Traversing graph is respectively as M (x) pixel value, width and height
The length and width of image, then utilize formula
Obtain average ave.Ave is related to the overall intensity with mist image, serves the effect of automatic adjusument brightness.
In order to allow recovery effect more to stablize, different scenes are adapted to, therefore introduce regulation parameter ρ and ρ ave.Through testing
Card,
Defog effect is best during ρ=C.In order to ensure that filtered result is compensated, be on the occasion of and can not be too small, take
0.9 is limited to, atmospheric scattering function is obtained
L (x)=min { min (ρ ave, 0.9) Mblur(x), M (7)
The image of atmospheric scattering function is as indicated with 6.
5. the specific steps tried to achieve fog free images by greasy weather Imaging physics model and exported include:
In computer vision, according to Mie scattering principle, greasy weather Imaging physics model can tabular form be:
I (x)=J (x) t (x)+A (1-t (x) (8)
Wherein I (x) is the foggy image observed;J (x) is the fog free images recovered;A represents global atmosphere light;t(x)
For atmospheric transmissivity, represent that the ratio for scattering and being transmitted directly to video camera does not occur for the light that object reflects in scene.From
Formula (8) is it can be seen that atmospherical scattering model is constituted by 2:Section 1 is incident light attenuation model, represents body surface reflection
The radiation intensity of light in atmosphere after propagation attenuation;Section 2 is atmospheric dissipation function, represents other light in atmospheric environment
To the influence produced by imaging, cause the skew of image color and brightness.Wherein, atmospheric dissipation function L (x) is also denoted as
L (x)=A (1-t (x)) (9)
Then (8) formula can be rewritten as
Estimated out from foggy image I (x) after global atmosphere light A and atmospheric dissipation function L (x), formula (10) deformation
For
Just clearly fog free images J (x) can be recovered and exported, it is fogless as shown in Figure 7.
Claims (3)
1. a kind of UUV Layer Near The Sea Surfaces visible images defogging method, it is characterized in that:
(1) original foggy image is obtained, the minimum value of three Color Channels is asked for, is designated as M (x);
(2) global atmosphere light is asked for using M (x) and quaternary tree close classification improved method;
(3) adaptive surface is carried out to M (x) to obscure;
(4) atmospheric scattering function is obtained;
(5) fog free images are tried to achieve by greasy weather Imaging physics model and exported.
2. UUV Layer Near The Sea Surfaces visible images defogging method according to claim 1, it is characterized in that the utilization M (x) and four
The improved method of fork tree close classification obtains global atmosphere light and specifically included:
1) image is divided into four rectangular areas;
2) average pixel value and variance in each region are asked for, average pixel value subtracts variance and is considered as each area score;
3) it is four zonules by the region division of highest scoring;
4) repeat step 2)~3), until the area in top score region is less than threshold value S set in advanceT;
5) the most bright spot in the region is taken, its pixel value is global atmosphere light A.
3. UUV Layer Near The Sea Surfaces visible images defogging method according to claim 2, it is characterized in that described carried out certainly to M (x)
Adapt to specifically including for surface blur:
Adaptive surface Fuzzy Processing has two parameters, radius s and threshold value T, and radius s determines fuzzy scope, and threshold value T determines mould
The degree of paste, fuzzy scope is exactly the size of convolution matrix, when radius is s, a diameter of 2s+1 of fuzzy matrix, matrix element
Number is s*s, and the neutral element of matrix is current pixel value, and the calculation formula of matrix element value is:
<mrow>
<msub>
<mi>w</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msup>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>M</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>-</mo>
<msub>
<mi>M</mi>
<mn>0</mn>
</msub>
</mrow>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mi>T</mi>
</mfrac>
</mrow>
Wherein T is threshold value, wijFor the element value of pattern matrix, be also weight, MijFor image M (x) pixel values, M0For pattern matrix
The pixel value at center;
Then to wijPre-process
wij=max (1, wij)
According to convolution algorithm, each pixel is by the value after obscuring:
<mrow>
<msub>
<mi>M</mi>
<mrow>
<mi>b</mi>
<mi>l</mi>
<mi>u</mi>
<mi>r</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<msub>
<mi>&Sigma;w</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<msub>
<mi>M</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
</mrow>
<mrow>
<msub>
<mi>&Sigma;w</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
</mrow>
</mfrac>
</mrow>
As the pixel value M at pattern matrix center0It is bright areas, blur radius s=3, window size is 7*7 during > A-50;
As the pixel value M at pattern matrix center0During≤A-50, for local dark areas, windows radius s=1, using 3*3 window
Mouthful.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710362665.0A CN107248146B (en) | 2017-05-22 | 2017-05-22 | Defogging method for UUV visible light image on offshore surface |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710362665.0A CN107248146B (en) | 2017-05-22 | 2017-05-22 | Defogging method for UUV visible light image on offshore surface |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107248146A true CN107248146A (en) | 2017-10-13 |
CN107248146B CN107248146B (en) | 2020-09-11 |
Family
ID=60017421
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710362665.0A Active CN107248146B (en) | 2017-05-22 | 2017-05-22 | Defogging method for UUV visible light image on offshore surface |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107248146B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107808368A (en) * | 2017-11-30 | 2018-03-16 | 中国电子科技集团公司第三研究所 | A kind of color image defogging method under sky and ocean background |
CN112949389A (en) * | 2021-01-28 | 2021-06-11 | 西北工业大学 | Haze image target detection method based on improved target detection network |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101493939A (en) * | 2009-02-27 | 2009-07-29 | 西北工业大学 | Method for detecting cooked image based on small wave domain homomorphic filtering |
CN101901473A (en) * | 2009-05-31 | 2010-12-01 | 汉王科技股份有限公司 | Self-adaptive defogging strengthening method of single-frame image |
US20110135200A1 (en) * | 2009-12-04 | 2011-06-09 | Chao-Ho Chen | Method for determining if an input image is a foggy image, method for determining a foggy level of an input image and cleaning method for foggy images |
CN102938136A (en) * | 2012-07-19 | 2013-02-20 | 中国人民解放军国防科学技术大学 | Method for defogging single images based on Bayer formats rapidly |
CN104715445A (en) * | 2013-12-13 | 2015-06-17 | 腾讯科技(深圳)有限公司 | Image processing method and system |
CN105654440A (en) * | 2015-12-30 | 2016-06-08 | 首都师范大学 | Regression model-based fast single-image defogging algorithm and system |
CN105761230A (en) * | 2016-03-16 | 2016-07-13 | 西安电子科技大学 | Single image defogging method based on sky region segmentation processing |
CN105787904A (en) * | 2016-03-25 | 2016-07-20 | 桂林航天工业学院 | Adaptive global dark channel prior image dehazing method for bright area |
CN106204494A (en) * | 2016-07-15 | 2016-12-07 | 潍坊学院 | A kind of image defogging method comprising large area sky areas and system |
-
2017
- 2017-05-22 CN CN201710362665.0A patent/CN107248146B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101493939A (en) * | 2009-02-27 | 2009-07-29 | 西北工业大学 | Method for detecting cooked image based on small wave domain homomorphic filtering |
CN101901473A (en) * | 2009-05-31 | 2010-12-01 | 汉王科技股份有限公司 | Self-adaptive defogging strengthening method of single-frame image |
US20110135200A1 (en) * | 2009-12-04 | 2011-06-09 | Chao-Ho Chen | Method for determining if an input image is a foggy image, method for determining a foggy level of an input image and cleaning method for foggy images |
CN102938136A (en) * | 2012-07-19 | 2013-02-20 | 中国人民解放军国防科学技术大学 | Method for defogging single images based on Bayer formats rapidly |
CN104715445A (en) * | 2013-12-13 | 2015-06-17 | 腾讯科技(深圳)有限公司 | Image processing method and system |
CN105654440A (en) * | 2015-12-30 | 2016-06-08 | 首都师范大学 | Regression model-based fast single-image defogging algorithm and system |
CN105761230A (en) * | 2016-03-16 | 2016-07-13 | 西安电子科技大学 | Single image defogging method based on sky region segmentation processing |
CN105787904A (en) * | 2016-03-25 | 2016-07-20 | 桂林航天工业学院 | Adaptive global dark channel prior image dehazing method for bright area |
CN106204494A (en) * | 2016-07-15 | 2016-12-07 | 潍坊学院 | A kind of image defogging method comprising large area sky areas and system |
Non-Patent Citations (4)
Title |
---|
HUIYAN LIU 等: "An Improved Fog-degrading Image Enhancement Algorithm Based on the Fuzzy Contrast", 《2010 INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND SECURITY》 * |
JIN-HWAN KIM 等: "Optimized contrast enhancement for real-time image and video dehazing", 《J.VIS.COMMUN.IMAGE R》 * |
何文章 等: "基于模糊对比度的雾天图像清晰化算法", 《科技导报》 * |
吴迪 等: "图像去雾的最新研究进展", 《自动化学报》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107808368A (en) * | 2017-11-30 | 2018-03-16 | 中国电子科技集团公司第三研究所 | A kind of color image defogging method under sky and ocean background |
CN112949389A (en) * | 2021-01-28 | 2021-06-11 | 西北工业大学 | Haze image target detection method based on improved target detection network |
Also Published As
Publication number | Publication date |
---|---|
CN107248146B (en) | 2020-09-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106530246B (en) | Image defogging method and system based on dark Yu non local priori | |
CN106548463B (en) | Sea fog image automatic defogging method and system based on dark and Retinex | |
CN101901473B (en) | Self-adaptive defogging strengthening method of single-frame image | |
CN110211067B (en) | Defogging method for UUV visible light image on offshore surface | |
CN102750674B (en) | Video image defogging method based on self-adapting allowance | |
US9361670B2 (en) | Method and system for image haze removal based on hybrid dark channel prior | |
CN102831591B (en) | Gaussian filter-based real-time defogging method for single image | |
CN104933680B (en) | A kind of intelligent quick sea fog minimizing technology of unmanned boat vision system video | |
US20180122051A1 (en) | Method and device for image haze removal | |
CN107301623A (en) | A kind of traffic image defogging method split based on dark and image and system | |
CN104318524A (en) | Method, device and system for image enhancement based on YCbCr color space | |
CN109993804A (en) | A kind of road scene defogging method generating confrontation network based on condition | |
CN104182943B (en) | A kind of single image defogging method capable merging human-eye visual characteristic | |
CN103034983B (en) | A kind of defogging method capable based on anisotropic filtering | |
CN112614136A (en) | Infrared small target real-time instance segmentation method and device | |
CN107273903B (en) | UUV offshore visible light image sea-sky-line extraction method based on LSD improvement | |
CN103578083A (en) | Single image defogging method based on joint mean shift | |
Cheng et al. | Visibility enhancement of single hazy images using hybrid dark channel prior | |
Lu et al. | Single underwater image descattering and color correction | |
Mei et al. | Single image dehazing using dark channel fusion and haze density weight | |
CN107248146A (en) | A kind of UUV Layer Near The Sea Surfaces visible images defogging method | |
CN104809697B (en) | A kind of improved high quick sea fog image enchancing method of boostfiltering | |
CN106780362A (en) | Road video defogging method based on dichromatic reflection model Yu bilateral filtering | |
Raigonda et al. | Haze Removal Of Underwater Images Using Fusion Technique | |
Lee et al. | A 4K-capable hardware accelerator of haze removal algorithm using haze-relevant features |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | 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 |