CN109272475A - A kind of method of fast and effective reparation and reinforcing underwater picture color - Google Patents
A kind of method of fast and effective reparation and reinforcing underwater picture color Download PDFInfo
- Publication number
- CN109272475A CN109272475A CN201811007644.8A CN201811007644A CN109272475A CN 109272475 A CN109272475 A CN 109272475A CN 201811007644 A CN201811007644 A CN 201811007644A CN 109272475 A CN109272475 A CN 109272475A
- Authority
- CN
- China
- Prior art keywords
- image
- color
- lms
- original image
- transmittance
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration by the use of histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
Abstract
A kind of method of fast and effective reparation and reinforcing underwater picture color, in turn includes the following steps: original image being carried out progress white balance processing after color space conversion, is repaired to color of image；Image after repairing enhances image detail by carrying out image parameter adjusting after stretching to color channel histograms；Background colour weight protection processing is carried out to enhanced image；Using the parameter in treatment process in step (1)-(3), original image and revised image are handled according to the method for weighted sum, color offset phenomenon can be overcome, improves the clarity and processing speed of image.
Description
Technical field
The present invention relates to field of image processings, and in particular to a kind of side quickly and effectively repaired and strengthen underwater picture color
Method.
Background technique
Underwater Imaging has extensive and important answer in target detection, sea materials detection and ocean geography engineering under water
With value, the just pay attention to day by day by various countries researcher.For Underwater Imaging, back scattering is one for influencing image quality
Major reason, the influence for muddy water body back scattering just become all the more obvious.In common Underwater Imaging technology, use
Halogen lamp is as lighting source, and not only power consumption is big, and imaging effect is influenced seriously by water body scattering, in the situation of water body muddiness
Lower imaging effect is very poor.And laser brightness is high, good directionality, is the lighting source more paid close attention in recent years.General laser lighting
In conjunction with technological means such as range gating, laser line scannings, the influence of back scattering can be effectively reduced, however range gating method
At high cost, complicated for operation, the laser line scanning imaging time is long, and due to the shaking of water body, needs to carry out between image effectively
Correction.
According to underwater optics imaging model principle, the image of underwater photograph technical is due to by the underwater decaying of light and scattering
It influences, leads to showed pattern colour tuningout blue-green.And it is bright that particle, suspended matter in water etc. also result in image
The problems such as spot and image are fuzzy, and contrast is low, and brightness is partially dark.The image quality of underwater picture has been seriously affected, it also can be different
The influence of degree further uses other operations of underwater picture progress, such as feature extraction, target identification, image information statistics
Deng.
Presently, there are underwater picture color restorative procedure be mainly all based on dark channel prior theory, water after simplification
On the basis of lower imaging model, calculate the transmissivity and background colour of original image, then according to each channel color decay the case where into
Row compensation.Wherein in order to improve the contrast and brightness of image, can also be stretched further combined with white balance method and histogram
Method handles image.
Under the premise of water quality of the same race, exponential increase can be presented with the increase of the depth of water in the attenuation degree of light, and the depth of water reaches
To a certain degree, red channels almost decay to zero (or can ignore compared to blue light and green optical channel is zero) then not
Meet the application range of dark channel prior theory.And if original image shows a certain leading color, other colors on the whole
Ingredient proportion very little can not meet dark channel prior theory.So directlying adopt the underwater picture of dark channel prior theory
Reparation can only be more satisfactory as a result, not having popularity for that can obtain under certain specific conditions.
Particularly, now with the fast development of the various sci-tech products with camera function, for can quickly handle
The requirement of the method for high-definition image is higher and higher, thus need it is a kind of can quickly and effectively application method come repair underwater picture with
And strengthen image information.
Summary of the invention
It is quickly and effectively repaired and reinforcing underwater picture it is an object of the invention to overcome the deficiencies of the prior art and provide a kind of
The method of color can overcome color offset phenomenon, improve the clarity and processing speed of image.
The present invention provides a kind of methods quickly and effectively repaired and strengthen underwater picture color, successively include following step
It is rapid:
(1) white balance processing is carried out after original image being carried out color space conversion, color of image is repaired；
(2) for the image after repairing, by carrying out image parameter adjusting after stretching to color channel histograms, to image
Details is enhanced；
Underwater transmittance figure picture is estimated using improved dark channel prior algorithm；
The edge of objects in images is extracted according to image edge detection algorithm；
The binary picture of the foreground picture of binaryzation is obtained in conjunction with morphological image operating method, and by the ash of this transmittance figure
Angle value is as weight distribution map；
(3) background colour weight protection processing is carried out to enhanced image；
(4) using the parameter in treatment process in step (1)-(3), according to the method for weighted sum to original image and amendment
Image afterwards is handled.
Further, the step (1) further includes to the detection for having white " exposure " part in original image and filtering out, to original
Brightness value is more than the part of preset threshold without any processing in figure.
Further, the step (1) further includes pre-processing to original image, described to pre-process to original image,
Including tentatively denoising, size compression and/or gray scale are smooth.
Further, color space is l α β color space or CIE lab color space in the step (1).
Further, the step (1) specifically:
1) image of RGB color is transformed into the space LMS: xj(m, n)=Txyz,ijfi(m,n)；
2) space LMS is transformed into XYZ space: lj(m, n)=Tlms,ijxi(m,n)；
3) LMS Spatial elements are taken into log operations, then converts it to l α β color space: lk,j(m, n)=Tpca, ijllog,i(m,n)；
4) it has been put in l α β color space or CIE lab color space by translation α β axis or ab axis average point to (0,0)
The correction of pairs of white balance point；
Wherein, Txyz,TlmsAnd TpcaIt is constant matrices, corresponding following table (i, j), (m, n) represents the row of corresponding pixel
And column, fi(m, n) is the image function of RGB color, xj(m, n) is the image function in the space LMS, lj(m, n) is XYZ empty
Between image function, llog,i(m, n) is that LMS Spatial elements are taken to the image function after log operations, lk,j(m, n) is to be transformed into l
The image function of α β color space or CIE lab color space.
It further, be image parameter in the step (2) is contrast, one of brightness and saturation degree or a variety of.
Further, the step (2) further include:
1) the edge distribution figure of image is extracted using Boundary extracting algorithm；
2) opposite side fate Butut is handled to obtain binary image, and wherein white area represents non-water body background area,
Black region represents water body background area；
3) dark channel prior theoretical algorithm combination background colour information is used, the transmittance figure of image is calculated, to transmissivity
Figure is filtered.
Further, the hypothesis in the step 3) in dark channel prior theoretical algorithm are as follows:
Wherein, JnTo input each channel of BGR image, n B, G, R, y is on the section Ω (x) centered on pixel x
Pixel, " :=" indicate to be defined as.
Further, the step 3) in the step (2) specifically:
A. dark channel prior formula is utilized, transmittance figure is obtained, threshold value is [0,1]；
B. the gray level image of original image is obtained；
C. gray level image is combined to be filtered the transmittance figure of acquisition.
The filtering processing uses Steerable filter method or bilateral filtering method.
Further, the step (4) specifically:
Background area and non-background area are distinguished by transmittance figure, then retains the effect of non-background area color reparation
Water body background area is amplified to full size size by fruit, and using the method for weighted sum, the background colour of original image is incorporated everywhere
In the background colour in image after reason.
Further, the step (2) further include:
By channel each for hsv color carry out histogram stretching during, for pixel in histogram pixel
Value sequenceMore than max-thresholds percentage σmaxPart be taken as maximum value, it may be assumed that P0=Pmax,ForLess than minimum threshold percentage σminPart be taken as minimum value, it may be assumed that P0=Pmin,Histogram stretches
Transformation relation formula are as follows: P0=(Pi-Pmin)*(Vmax-Pmin)/(Pmax-Pmin)+Vmin。
The fast and effective method repaired and strengthen underwater picture color of the invention, may be implemented:
1) red channel information attenuation is serious in the case where imaging depth relatively depth, in color image (compares bluish-green chrominance channel
It is negligible), it will lead to serious color offset phenomenon if directlying adopt dark channel prior theoretical method and repairing original image
Problem, color image color distortion caused by the present invention solves the decaying of different imaging depths under normal circumstances and scatters have
Good effect overcomes serious color offset phenomenon.
2) pass through image enhancement technique, hence it is evident that improve the details of image, strengthen the contrast of image, improve original image
Partially dark problem, so that treated, image presents more image details, it appears that more realistic.
3) this method is improved in various application platforms (computer platform, embedded platform, mobile phone mobile terminal platform etc.)
The efficiency of realization.
Detailed description of the invention
Fig. 1 is the method flow diagram repaired with strengthen underwater picture color；
Fig. 2 is the method flow diagram quickly and effectively repaired with strengthen underwater picture color；
Fig. 3 is that RGB color is converted to l α β color space flow chart.
Specific embodiment
The following detailed description of specific implementation of the invention, it is necessary to it is indicated herein to be, implement to be only intended to this hair below
Bright further explanation, should not be understood as limiting the scope of the invention, and field person skilled in the art is according to above-mentioned
Some nonessential modifications and adaptations that summary of the invention makes the present invention, still fall within protection scope of the present invention.
The present invention provides a kind of method quickly and effectively repaired and strengthen underwater picture color, principle such as attached drawings 1, figure
Shown in 2 and Fig. 3, introduce further below.
Quickly and effectively repairing mainly includes color of image reparation with the method for strengthening underwater picture color, and image detail increases
The step of strong and image background color weight is protected (as shown in Figure 1), specific:
Color of image reparation is specifically, carrying out color white balance processing to original image in l α β color space (also can be used
CIE lab color space substitution), repair serious channel of decaying.The Processing Algorithm is to assume (gray- based on gray world
World assumption) realize, mean value of the immersed body to the reflection of light source be assumed to gray value, so under water at
The blue-green showed on the whole is eliminated in the image of picture to be influenced.
Its detailed process are as follows:
1) image of RGB color is transformed into the space LMS: x firstj(m, n)=Txyz,ijfi(m,n)；
2) space LMS is transformed into XYZ space: lj(m, n)=Tlms,ijxi(m,n)；
3) LMS Spatial elements are taken into log operations, i.e. log (LMS) then converts it to l α β color space:
llαβ,j(m, n)=Tpca,ijllog,i(m,n)。
Wherein, Txyz,TlmsAnd TpcaIt is constant matrices, corresponding subscript (i, j), (m, n) represents the row of corresponding pixel
And column, fi(m, n) is the image function of RGB color, xj(m, n) is the image function in the space LMS, lj(m, n) is XYZ empty
Between image function, llog,i(m, n) is that LMS Spatial elements are gone to the image function after log operations, llαβ,j(m, n) is transformed into l
The image function of α β color space.
4) correction to white balance point is completed by translation α β axis average point to (0,0) point in l α β color space,
That is:WhereinFor the image function after white balance correction,For each channel
Average pixel value.
After being repaired by color, the step of enhancing for image detail, by from RGB, hsv color channel selectivity
To the contrast of image, saturation degree is adjusted, and obtained image totally eliminates the color of blueing green tone.And pass through
After histogram stretching conversion, the contrast of image, brightness has all obtained relatively good adjusting, wherein HSL color also can be used
Space replaces hsv color space.
By edge detection method, the marginal information of each imaging object in detection image passes through the operation of morphological image
Mode divides an image into the module distribution figure of binaryzation, and selects the large area block of not limbus, and selection meets one
The maximum region block of fixed condition, and using the region as background area, obtain background colour information.
Caused by the color that underwater picture background is presented is primarily due to decaying, attenuation degree and the depth of water, water quality,
The many factors such as geographical location are related, so water body background area only cannot be obtained by the way that the distribution of color is imaged.But it examines
Consider water body background area in image, on the whole image gradient distribution it is more uniform, other regions in image mostly have compared with
For apparent Edge texture information, and this feature is relatively stable by influence of fading, therefore uses canny Boundary extracting algorithm, the calculation
While method inhibits noise, object marginal information can be retained to greatest extent.In the Edge texture point for obtaining image by canny
After cloth, operated by image " expansion ", the region connection with edge feature is blocking, it is operated finally by image " corrosion "
The edge of contracting pocket, the edge distribution trend in projecting edge region.Both available similar to two-value diagram form in this way
Underwater background area weight distribution map, it should be noted that Canny Edge Detection can pass through Sobel Edge Detection
Substitution can be used gradient map by edge detection method and be distributed detection method substitution.
Detailed process are as follows:
1) the edge distribution figure of image is extracted using canny Boundary extracting algorithm.
2) opposite side fate Butut carries out image " expansion " operation.
3) image " corrosion " operation is carried out to the figure after image " expansion " operation.Binary image is obtained, wherein white area
Domain represents non-water body background area, and black region represents water body background area.
It is theoretical (Dark Channel Prior, DCP) using improved dark channel prior, in conjunction with background colour information, calculate
The transmittance figure of image out.
Compared to attenuation model t (x)=e of the light in mist-βd(x)With attenuation model t (x, λ)=e in water-c(λ)d(x)It is not difficult
It was found that the attenuation coefficient c (λ) of different wave length is related to wavelength X, it is inconsistent that the light guide of different wave length causes light to decay under water.Especially
Ground, in the waters that the depth of water is deeper, red channel decaying is serious, will have a direct impact on the estimation of dark, therefore by former dark channel prior
Hypothesis in theory:
It is changed to
Wherein, JnTo input each channel of BGR image, n B, G, R, y is on the section Ω (x) centered on pixel x
Pixel." :=" expression " being defined as ", for defining an emerging symbol.
To realize that the dark to underwater different depth image is estimated.The dark channel image that Steerable filter algorithm will acquire
It is filtered operation with the grayscale image of original image, the noise in smooth original dark channel diagram remains the edge of grayscale image.
Concrete operations process are as follows:
1) using dark channel prior formula is improved, dark channel image is obtained；
2) gray level image of original image is obtained；
3) operation is filtered using dark channel image and original image grayscale image of the Steerable filter algorithm to acquisition.
Transmittance figure is filtered using Steerable filter method (Guided Filter), is calculated according to Steerable filter
Method is filtered operation in conjunction with grayscale image.
Background area and non-background area are distinguished by transmittance figure, then retains the effect of non-background area color reparation
Fruit will be in the background colour in the background colour of original image image of being dissolved into that treated in the way of weighted sum.
By weighted summationw(x,y)∈[0,
1], by original image IsrcImage I after (x, y) and color correctcor(x, y) is overlapped, wherein in transmittance figure w each pixel
The value of point can regard the weight of the pixel as corresponding original image and revised figure as, wherein (x, y) indicates that pixel is being schemed
Coordinate as in, Ω indicate background area.
Its detailed process are as follows:
1) by the binary picture obtained, the transmittance figure of background water area is partitioned into from transmittance figure；
2) original image and revised image are calculated according to weighted sum formula.
In order to improve the operational efficiency of practical application this method processing high-definition image, propose a kind of using analysis and processing point
The mode opened, analysis include obtaining the calculated parameter of institute in color of image repair process, are counted in image enhancement processes
The calculated parameter of institute during the parameter of calculating and the weight distribution map of non-background protection.Processing includes after getting parms, directly
It connects and high definition original image is operated, reduce the operation times of direct operation high definition original image, and then this method can be improved in reality
Execution speed in.
Its detailed process is
1) original image is compressed, so that compressed size is no more than or approximation 320p image size, processing 4K figure opposite in this way
As the time can reduce the 1% of about original place reason time.
2) compressed image is operated according to the white balance method in above-mentioned steps, and is retained in calculating process
Parameter.
3) according to the contrast adjustment method in above-mentioned steps, to image degree of the comparing tune by white balance processing
Section, and retain the parameter in calculating process.
4) according to the water body background area extracting method in above-mentioned steps, the background area of compressed original image is obtained.
5) according to the improvement dark channel prior method in above-mentioned steps, the transmittance figure of compressed original image is obtained.
6) to the saturating color rate figure obtained in above-mentioned steps, Steerable filter behaviour is carried out according to the filtering mode in above-mentioned steps
Make, is may be otherwise certainly using bilateral filtering method.
7) step 2) -4 is combined) parameter in operation write and realizes above step using openGL shader language
Shader processing routine utilizes GPU high speed processing original image.
8) water body background area is amplified to full size size, according to the weighted sum mode in above-mentioned steps, to original image
Carry out final weighted sum processing.
Although for illustrative purposes, it has been described that exemplary embodiments of the present invention, those skilled in the art
Member it will be understood that, can be in form and details in the case where the scope and spirit for not departing from invention disclosed in appended claims
On the change that carry out various modifications, add and replace etc., and all these changes all should belong to appended claims of the present invention
Protection scope, and each step in the claimed each department of product and method, can in any combination
Form is combined.Therefore, to disclosed in this invention the description of embodiment be not intended to limit the scope of the invention,
But for describing the present invention.Correspondingly, the scope of the present invention is not limited by embodiment of above, but by claim or
Its equivalent is defined.
Claims (11)
1. a kind of method quickly and effectively repaired and strengthen underwater picture color, which is characterized in that in turn include the following steps:
(1) white balance processing is carried out after original image being carried out color space conversion, color of image is repaired；
(2) for the image after repairing, by carrying out image parameter adjusting after stretching to color channel histograms, to image detail
Enhanced；
Underwater transmittance figure picture is estimated using improved dark channel prior algorithm；
The edge of objects in images is extracted according to image edge detection algorithm；
The binary picture of the foreground picture of binaryzation is obtained in conjunction with morphological image operating method, and by the gray value of this transmittance figure
As weight distribution map；
(3) background colour weight protection processing is carried out to enhanced image；
(4) using the parameter in treatment process in step (1)-(3), according to the method for weighted sum to original image and revised
Image is handled.
2. the method as described in claim 1, it is characterised in that: the step (1) further include to have in original image white " expose
The detection of light " part with filter out, to brightness value in original image be more than preset threshold part without any processing.
3. the method as described in claim 1, it is characterised in that: the step (1) further includes pre-processing to original image, institute
Stating and carrying out pretreatment to original image includes preliminary denoising, and size compression and/or gray scale are smooth.
4. method as claimed in claim 3, it is characterised in that: in the step (1) color space be l α β color space or
CIE lab color space.
5. method as claimed in claim 4, it is characterised in that: the step (1) specifically:
1) image of RGB color is transformed into the space LMS: xj(m, n)=Txyz,ijfi(m,n)；
2) space LMS is transformed into XYZ space: lj(m, n)=Tlms,ijxi(m,n)；
3) LMS Spatial elements are taken into log operations, then converts it to l α β color space: lk,j(m, n)=Tpca,ijllog,i
(m,n)；
4) completion pair is put by translation α β axis or ab axis average point to (0,0) in l α β color space or CIE lab color space
The correction of white balance point；
Wherein, Txyz,TlmsAnd TpcaBe constant matrices, corresponding following table (i, j), (m, n) represent corresponding pixel row and
Column, fi(m, n) is the image function of RGB color, xj(m, n) is the image function in the space LMS, lj(m, n) is XYZ space
Image function, llog,i(m, n) is that LMS Spatial elements are taken to the image function after log operations, lk,j(m, n) is to be transformed into l α β
The image function of color space or CIE lab color space.
6. the method as described in claim 1, it is characterised in that: be image parameter in the step (2) be contrast, brightness and
One of saturation degree is a variety of.
7. such as claim 1-2, the described in any item methods of 4-6, it is characterised in that: the step (2) further include:
1) the edge distribution figure of image is extracted using Boundary extracting algorithm；
2) opposite side fate Butut is handled to obtain binary image, and wherein white area represents non-water body background area, black
Regional Representative's water body background area；
3) use dark channel prior theoretical algorithm combination background colour information, calculate the transmittance figure of image, to transmittance figure into
Row filtering processing.
8. the method for claim 7, it is characterised in that: the vacation in the step 3) in dark channel prior theoretical algorithm
It is set as:
Wherein, JnTo input each channel of BGR image, n B, G, R, y is the pixel on the section Ω (x) centered on pixel x
Point, " :=" indicate to be defined as.
9. method according to claim 8, it is characterised in that: the step 3) in the step (2) specifically:
A. dark channel prior formula is utilized, transmittance figure is obtained, threshold value is [0,1]；
B. the gray level image of original image is obtained；
C. gray level image is combined to be filtered the transmittance figure of acquisition.
The filtering processing uses Steerable filter method or bilateral filtering method.
10. such as claim 1-2,4-6,8, method described in any one of 9, it is characterised in that: the step (4) specifically:
Background area and non-background area are distinguished by transmittance figure, then retains the effect of non-background area color reparation, it will
Water body background area is amplified to full size size, using the method for weighted sum, after the background colour of original image is dissolved into processing
Image in background colour in.
11. method as described in claim 1, it is characterised in that: the step (2) further include:
By channel each for color carry out histogram stretching during, for pixel in histogram rank-ordered pixels
Pi indexMore than max-thresholds percentage σmaxPart be taken as maximum value, it may be assumed that P0=Pmax,Pi index∈[σmax, 1], for
Pi indexLess than minimum threshold percentage σminPart be taken as minimum value, it may be assumed that P0=Pmin,Pi index∈[0,σmin], histogram Tula
Stretch transformation relation formula are as follows: P0=(Pi-Pmin)*(Vmax-Pmin)/(Pmax-Pmin)+Vmin。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811007644.8A CN109272475B (en) | 2018-08-31 | 2018-08-31 | Method for rapidly and effectively repairing and strengthening underwater image color |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811007644.8A CN109272475B (en) | 2018-08-31 | 2018-08-31 | Method for rapidly and effectively repairing and strengthening underwater image color |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109272475A true CN109272475A (en) | 2019-01-25 |
| CN109272475B CN109272475B (en) | 2022-02-18 |
Family
ID=65154792
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811007644.8A Active CN109272475B (en) | 2018-08-31 | 2018-08-31 | Method for rapidly and effectively repairing and strengthening underwater image color |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109272475B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110570365A (en) * | 2019-08-06 | 2019-12-13 | 西安电子科技大学 | image defogging method based on prior information |
| CN111626965A (en) * | 2020-06-04 | 2020-09-04 | 成都星时代宇航科技有限公司 | Remote sensing image processing method, device, electronic equipment and storage medium |
| CN113920927A (en) * | 2021-10-25 | 2022-01-11 | 武汉华星光电半导体显示技术有限公司 | Display method, display panel and electronic equipment |
| CN113923474A (en) * | 2021-09-29 | 2022-01-11 | 北京百度网讯科技有限公司 | Video frame processing method and device, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104166968A (en) * | 2014-08-25 | 2014-11-26 | 广东欧珀移动通信有限公司 | Image dehazing method and device and mobile terminal |
| CN104766284A (en) * | 2015-04-17 | 2015-07-08 | 河海大学常州校区 | Underwater color blurred image restoration method |
| CN106023110A (en) * | 2016-05-20 | 2016-10-12 | 河海大学 | Image defogging method with high fidelity |
| CN106485681A (en) * | 2016-10-18 | 2017-03-08 | 河海大学常州校区 | Color image restoration method under water based on color correction and red channel prior |
| CN107067386A (en) * | 2017-04-24 | 2017-08-18 | 上海海洋大学 | A kind of shallow sea underwater picture Enhancement Method stretched based on relative color histogram |
| CN107798665A (en) * | 2017-11-07 | 2018-03-13 | 天津大学 | Underwater picture Enhancement Method based on structural texture layering |
| CN108269242A (en) * | 2018-01-17 | 2018-07-10 | 深圳市华星光电半导体显示技术有限公司 | Image enchancing method |
-
2018
- 2018-08-31 CN CN201811007644.8A patent/CN109272475B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104166968A (en) * | 2014-08-25 | 2014-11-26 | 广东欧珀移动通信有限公司 | Image dehazing method and device and mobile terminal |
| CN104766284A (en) * | 2015-04-17 | 2015-07-08 | 河海大学常州校区 | Underwater color blurred image restoration method |
| CN106023110A (en) * | 2016-05-20 | 2016-10-12 | 河海大学 | Image defogging method with high fidelity |
| CN106485681A (en) * | 2016-10-18 | 2017-03-08 | 河海大学常州校区 | Color image restoration method under water based on color correction and red channel prior |
| CN107067386A (en) * | 2017-04-24 | 2017-08-18 | 上海海洋大学 | A kind of shallow sea underwater picture Enhancement Method stretched based on relative color histogram |
| CN107798665A (en) * | 2017-11-07 | 2018-03-13 | 天津大学 | Underwater picture Enhancement Method based on structural texture layering |
| CN108269242A (en) * | 2018-01-17 | 2018-07-10 | 深圳市华星光电半导体显示技术有限公司 | Image enchancing method |
Non-Patent Citations (2)
| Title |
|---|
| XIU LI ET AL.: "Underwater image enhancement via dark channel prior and luminance adjustment", 《OCEANS 2016 - SHANGHAI》 * |
| 朱振杰 等: "基于暗原色先验原理的水下图像增强", 《图学学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110570365A (en) * | 2019-08-06 | 2019-12-13 | 西安电子科技大学 | image defogging method based on prior information |
| CN110570365B (en) * | 2019-08-06 | 2022-04-29 | 西安电子科技大学 | Image defogging method based on prior information |
| CN111626965A (en) * | 2020-06-04 | 2020-09-04 | 成都星时代宇航科技有限公司 | Remote sensing image processing method, device, electronic equipment and storage medium |
| CN111626965B (en) * | 2020-06-04 | 2021-03-16 | 成都星时代宇航科技有限公司 | Remote sensing image processing method, device, electronic equipment and storage medium |
| CN113923474A (en) * | 2021-09-29 | 2022-01-11 | 北京百度网讯科技有限公司 | Video frame processing method and device, electronic equipment and storage medium |
| WO2023050829A1 (en) * | 2021-09-29 | 2023-04-06 | 北京百度网讯科技有限公司 | Video frame processing method and apparatus, and electronic device and storage medium |
| CN113920927A (en) * | 2021-10-25 | 2022-01-11 | 武汉华星光电半导体显示技术有限公司 | Display method, display panel and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109272475B (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105761227B (en) | Underwater picture Enhancement Method based on dark channel prior and white balance | |
| CN109272475A (en) | A kind of method of fast and effective reparation and reinforcing underwater picture color | |
| US20050175253A1 (en) | Method for producing cloud free and cloud-shadow free images | |
| Pei et al. | Effective image haze removal using dark channel prior and post-processing | |
| CN109447917B (en) | Remote sensing image haze eliminating method based on content, characteristics and multi-scale model | |
| CN108182671B (en) | Single image defogging method based on sky area identification | |
| CN110827225A (en) | Non-uniform illumination underwater image enhancement method based on double exposure frame | |
| CN110930321A (en) | Blue/green screen digital image matting method capable of automatically selecting target area | |
| Li et al. | Fast region-adaptive defogging and enhancement for outdoor images containing sky | |
| Nnolim | Image de-hazing via gradient optimized adaptive forward-reverse flow-based partial differential equation | |
| Parihar et al. | A comprehensive analysis of fusion-based image enhancement techniques | |
| CN115456905A (en) | Single image defogging method based on bright and dark region segmentation | |
| CN108711160A (en) | A kind of Target Segmentation method based on HSI enhancement models | |
| Lei et al. | A novel intelligent underwater image enhancement method via color correction and contrast stretching | |
| Raigonda et al. | Haze Removal Of Underwater Images Using Fusion Technique | |
| Dhanya et al. | L-CLAHE intensification filter (L-CIF) algorithm for underwater image enhancement and colour restoration | |
| CN115170415A (en) | Low-illumination image enhancement method and system and readable storage medium | |
| de Dravo et al. | An adaptive combination of dark and bright channel priors for single image dehazing | |
| CN113012079A (en) | Low-brightness vehicle bottom image enhancement method and device and storage medium | |
| CN112950484A (en) | Method for removing color pollution of photographic image | |
| Yang et al. | An underwater image enhancement method based on color correction and local unsharp masking | |
| He et al. | Underwater image enhancement based on light attenuation prior and dual-image multi-scale fusion | |
| CN114331873B (en) | Non-uniform illumination color image correction method based on region division | |
| Mukaida et al. | Low-Light Image Enhancement Method Using a Modified Gamma Transform and Gamma Filtering-Based Histogram Specification for Convex Combination Coefficients | |
| Xu et al. | Model for Single Image Enhancement Based on Adaptive Contrast Stretch and Multiband Fusion |
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 | ||
| CP03 | Change of name, title or address |
Address after: 518000 806, block B, Jiuzhou electric appliance building, No. 007, Keji South 12th Road, high tech Zone, Yuehai street, Nanshan District, Shenzhen, Guangdong Patentee after: Shenzhen Yidong Blue Technology Co.,Ltd. Address before: 518000 room 209, building 17, maker Town, No. 1201 Liuxian Avenue, Taoyuan Street, Nanshan District, Shenzhen, Guangdong Patentee before: SHENZHEN NAVA TECHNOLOGY Co.,Ltd. |
|
| CP03 | Change of name, title or address |