CN115937032A - Underwater dark image restoration method and system - Google Patents
Underwater dark image restoration method and system Download PDFInfo
- Publication number
- CN115937032A CN115937032A CN202211632243.8A CN202211632243A CN115937032A CN 115937032 A CN115937032 A CN 115937032A CN 202211632243 A CN202211632243 A CN 202211632243A CN 115937032 A CN115937032 A CN 115937032A
- Authority
- CN
- China
- Prior art keywords
- image
- underwater
- dark image
- calculating
- scene
- 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.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Image Processing (AREA)
Abstract
The invention provides an underwater dark image restoration method and system, which relate to the technical field of image processing, and the specific scheme comprises the following steps: calculating a background light value and a scene depth according to prior information of the underwater dark image; based on the background light value and the scene depth, combining an underwater scene decomposition model, constructing an objective function and solving, and calculating a reflectivity image, an illumination image and a noise item; correcting the reflectivity image and the illumination image through contrast stretching and brightness enhancement to obtain a restored underwater dark image; the method is based on the static parameter optimization and scene decomposition model, has better restoration effect on underwater dark images with different degrees, and can effectively smooth image noise and restore image colors in the process of improving the image brightness.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to an underwater dark image restoration method and system.
Background
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
Human beings are crossing the information age, and images as important carriers of information play an important role in daily life and work; the underwater image is used as a carrier of ocean information and is an important way for human beings to explore the ocean information; the underwater image is widely applied to the fields of marine energy exploration, underwater rescue, marine ecological protection, marine organism research and the like.
As the underwater detection depth is increased, the underwater illumination becomes weaker, and an image acquired under a weak illumination condition (hereinafter referred to as an underwater dark image) generally shows defects of dim illumination, color degradation, loss of details, noise and the like. However, because the traditional underwater imaging model does not consider the brightness difference and image noise of different areas in the underwater scene, when the traditional underwater imaging model is used for restoring the underwater dark image, two problems exist: the restored underwater dark image has low contrast, dark color and larger noise gradient; the parameters are usually set by an empirical method, and the optimization is difficult to ensure.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides the method and the system for restoring the underwater dark image, which have better restoration effects on the underwater dark images with different degrees based on the static parameter optimization and the scene decomposition model, and can effectively smooth the image noise and restore the image color in the process of improving the image brightness.
In order to achieve the above object, one or more embodiments of the present invention provide the following technical solutions:
the first aspect of the invention provides an underwater dark image restoration method;
an underwater dark image restoration method, comprising:
calculating a background light value and a scene depth according to prior information of the underwater dark image;
based on the background light value and the scene depth, combining an underwater scene decomposition model, constructing an objective function and solving, and calculating a reflectivity image, an illumination image and a noise item;
and correcting the reflectivity image and the illumination image through contrast stretching and brightness enhancement to obtain a restored underwater dark image.
Further, the prior information of the underwater dark image includes: and prior information among underwater dark image channels and prior information of image color and saturation.
Further, the calculation method of the background light value is as follows:
step (1): averagely dividing the underwater dark image into a plurality of regions, and calculating the score of each region by adopting an evaluation function;
step (2): selecting a region with the highest score as a candidate region, repeating the step (1) until the size of the candidate region is smaller than a preset threshold value, and outputting the region;
and (3): and calculating the pixel average value of the output area as the background light value.
Further, the underwater scene decomposition model specifically includes:
P c (x)＝G c [1-s c (x)]+F c (x)·I c (x)·s c (x)+Q c (x)
wherein, P c (x) For underwater dark images, G c Is the value of background light, s c (x) As depth of scene, F c (x) As a reflectance image, I c (x) As an illumination image, Q c (x) Representing the noise term after illumination and scene depth attenuation, x is the pixel coordinate, c epsilon { r, g, b } is the red, green and blue three channels of the image, and ". The" represents point-to-point multiplication.
Further, the objective function is specifically:
wherein, the first and the second end of the pipe are connected with each other,
holds the term for unknown quantity>
Is a noise suppression term>
A hold term representing a luminance component>
For a smooth term of reflectivity>
Represents a gradient filtering template in a four-neighbor domain, | |. | non-woven phosphor 2 Andi 1 Is L2 and L1 norm, respectively>
Representing convolution operations and I' representing a smooth underwater dark image. Phi is a 1 、φ 2 、φ 3 Are three parameters that balance the weights of the terms in the objective function.
Further, the restored underwater dark image specifically includes:
wherein, the first and the second end of the pipe are connected with each other,
is an illuminance map after brightness correction>
Is a reflectivity image after contrast stretching, x is a pixel coordinate, c belongs to { r, and } is a red channel, a green channel and a blue channel of the image, and' represents point-to-point multiplication.
Further, the method comprises the step of obtaining optimal parameters through a static parameter optimization method, wherein the method specifically comprises the following steps:
initializing a sample set;
calculating an image quality index of the sample;
screening out an excellent sample set according to the image quality indexes of all samples;
performing binary code exchange operation on the excellent sample set;
performing binary coding mutation operation on samples in the excellent sample set one by one;
and taking the sample with the largest image quality index in the excellent sample set as the optimal parameter.
The invention provides an underwater dark image restoration system in a second aspect.
An underwater dark image restoration system comprises a priori calculation module, a parameter optimization module, a function solving module and an image restoration module:
an a priori computation module configured to: calculating a background light value and a scene depth according to prior information of the underwater dark image;
a parameter optimization module configured to: obtaining optimal parameters by a static parameter optimization method;
a function solving module configured to: based on the background light value, the scene depth and the optimal parameters, combining an underwater scene decomposition model, constructing an objective function and solving, and calculating a reflectivity image, an illumination image and a noise item;
an image restoration module configured to: and correcting the reflectivity image and the illumination image through contrast stretching and brightness enhancement to obtain a restored underwater dark image.
A third aspect of the present invention provides a computer readable storage medium having stored thereon a program which, when executed by a processor, performs the steps in a method for underwater dark image restoration according to the first aspect of the present invention.
A fourth aspect of the present invention provides an electronic device, comprising a memory, a processor and a program stored on the memory and executable on the processor, wherein the processor implements the steps of the method for restoring an underwater dark image according to the first aspect of the present invention when executing the program.
The above one or more technical solutions have the following beneficial effects:
the invention provides a scene decomposition model by considering the difference of incident light intensity between a shadow area and a light direct-irradiating area during underwater imaging, decomposing a real scene in a traditional underwater imaging model into the product of illumination and reflectivity and considering a noise item in the reflectivity, solving the problem of color distortion caused by restored illumination, improving the color of a restored image, effectively smoothing image noise in the process of improving the image brightness, restoring the image color and having better restoration effect on underwater dark images with different degrees.
Aiming at the problem that parameters in the underwater dark image restoration method are set according to experience and the optimization cannot be guaranteed, the invention provides a static parameter optimization method which optimizes five parameters in an objective function and a linear mapping formula so as to obtain the optimal parameters and improve the restoration quality of the underwater dark image.
Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of the method of the first embodiment.
Fig. 2 is a graph showing the effect of the experiment of the first embodiment.
Fig. 3 is a system configuration diagram of the second embodiment.
Detailed Description
The invention is further described with reference to the following figures and examples.
Example one
The embodiment discloses an underwater dark image restoration method;
as shown in fig. 1, an underwater dark image restoration method includes:
step S1: calculating a background light value and a scene depth according to prior information of an underwater dark image, and specifically comprising the following steps:
step S101: and calculating the background light value of the underwater dark image according to the prior information among the underwater dark image channels.
The standard deviation of pixels in a background area of an underwater dark image is smaller than that in a foreground area, the difference values of a blue channel, a green channel and a red channel in the background area are larger than that in the foreground area, in addition, pixels representing background light values are usually located in a highlight area, and according to prior information among the underwater dark image channels, an evaluation function is innovatively constructed and used for calculating the value of the area:
where j denotes the jth region of the image, x is the pixel coordinate,
and &>
Is the pixel value of the pixel x in the jth area on the three channels red, green and blue respectively, and/or is selected>
Represents the average of the pixels in the jth region, k is the number of pixels in each region, c e { r, g, b } is three channels of red, green, and blue of the image, and ". Sup." represents a point-to-point multiplication. />
Based on the evaluation function, a background light value calculation method is provided as follows:
(1) Averagely dividing the underwater dark image into j regions, wherein j is set to be 8 in the embodiment, and calculating the score of each region by adopting an evaluation function;
(2) Selecting a region with the highest score, repeating the step (1) on the region with the highest score until the size of the region is smaller than a preset threshold value, wherein the size of the region is set to be 40 x 40, and outputting the region;
(3) Calculating the pixel mean value of the output area as the background light value G c Is a three-dimensional array.
Step S102: and calculating the scene depth of the underwater dark image according to the color and saturation prior information of the image.
Specifically, a saturation prior is added in a red channel prior, and then a transmission map is calculated by utilizing the red channel prior to prevent excessive enhancement, so that the scene depth s c () The calculation method is as follows:
wherein T (x) represents a saturation prior map,
a minimum filter with a filtering window of 5 × 5; r 、 g 、 b respectively representing the background light values of red, green and blue channels, collectively referred to as c Where c ∈ { r, }; r ()、 g ()、 b () Red, green and blue channels respectively representing dark underwater images, collectively called c () Where c ∈ { r, g, b }; r () A red channel for the calculated scene depth; according to the calculated r () Can be calculated out g () And b ()：
wherein the content of the first and second substances, g () And b () A green and a blue channel, respectively, of the scene depth, r ()、 g ()、 b () Are collectively referred to as c ()，c∈{r,g,b}。
Step S2: and constructing an objective function and solving the objective function based on the background light value and the scene depth in combination with the underwater scene decomposition model, and calculating a reflectivity image, an illumination image and a noise item.
The underwater image restoration method usually adopts a conventional underwater imaging model to enhance the contrast of the image, and the conventional underwater imaging model is as shown in formula (6):
P c (x)＝G c ·[1-s c (x)]+Z c (x)·s c (x) c∈{r, g, b} (6)
wherein, the first and the second end of the pipe are connected with each other, c () Is a dark image under water and is a dark image, c as a value of the background light, the ambient light, c () To be the depth of the scene, c () For a real scene, x is the pixel coordinate, c ∈ { r, } is the red, green and blue channels of the image, and ". Represents point-to-point multiplication.
The traditional underwater imaging model does not consider the difference of incident light intensity between a shadow area and a direct light area, so that the color of a restored image is darker, and color distortion caused by illumination cannot be restored, so that a real underwater dark image is difficult to restore; to this problem, the present embodiment innovatively proposes: real scene Z in traditional underwater imaging model c () Decomposed into illumination I c () And a reflectivity F c () And taking into account the noise term q in the reflectivity c () Expressed as:
Z c (x)＝[F c (x)+q c (x)]·I c (x) (7)
wherein, F c () As a reflectance image, q c () In order to be noise in the reflectivity image, c () Is an illumination image.
Substituting equation (7) into equation (6) yields:
P c (x)＝G c [1-s c (x)]+[F c (x)+q c (x)]·I c (x)·s c (x) (8)
let Q c ()＝q c ()·s c ()·I c () The following can be obtained:
P c (x)＝G c [1-s c (x)]+F c (x)·I c (x)·s c (x)+Q c (x) (9)
wherein, the first and the second end of the pipe are connected with each other, c () Representing the noise term attenuated by illumination and scene depth.
Formula (9) is the proposed underwater scene decomposition model, in which only P is present c () Is a known quantity, the background light value c Depth of scene c () The reflectance image F, which has been calculated in step S1 c () Noise Q in reflectance image c () Illumination image I c () Is an unknown quantity.
To solve F c ()、Q c () And c () In this embodiment, an objective function is innovatively constructed according to the underwater scene decomposition model of formula (9), as shown in formula (10):
wherein G, F, I, s, Q, P are each G c 、F c (x)、I c (x)、s c (x)、Q c (x)、P c (x) For short.
And data holding items of the three unknown quantities of F, I and Q are used for limiting the value ranges of the three unknown quantities of F, I and Q so as to meet the constraint of the underwater scene decomposition model. />
Is a noise suppression term>
A hold term representing a luminance component>
For a smooth term of reflectivity>
Represents a gradient filtering template in a four-neighbor domain, | |. | non-woven phosphor 2 Andi 1 Represents respectively L2 and L1 norm, <' > or>
Representing convolution operations, I' representing a smooth underwater dark image, phi 1 、φ 2 、φ 3 Three parameters are used to balance the weights of the terms in the objective function.
The value range of the pixel value in F is [0,1], so the constraint condition of F is:
0≤F≤1 (11)
substituting equation (11) into equation (9) can obtain the constraint condition of the illumination image I:
calculating the unknown quantity in the objective function by adopting an alternative direction multiplier method, wherein the calculation process comprises the following steps:
step S201: the derivative of the term in the objective function independent of Q with respect to Q is zero in equation (10), which is specifically:
deleting the terms in equation (10) that do not contain Q, resulting in:
equation (13) derives Q and let the derivative be 0, obtaining the optimal solution for Q:
step S202: the derivative of the term independent of I in the objective function of equation (10) to I is zero, specifically:
delete the I-independent term in equation (10) and will
Is transformed into
The following can be obtained:
equation (15) derives I and let the derivative be 0, which yields the optimal solution for I:
where z represents the result of the z-th iteration and ep is an infinitesimal positive number to ensure that the denominator is not 0.
Step S203: delete the F-independent term in equation (10) and will
Equivalent transformation to->
The following can be obtained:
formula (17) is an L1 norm optimization problem, and in the z-th iteration process, the optimal solution of F is:
where d iv is the divergence and σ is the time constant, this patent is set to 0.249.
Step S204: and (3) circularly and iteratively executing the steps S201, S202 and S203 until convergence is reached, and executing constraint conditions I which are greater than or equal to G + [ P-G-Q ]/S of I and F and are greater than or equal to 0 and less than or equal to 1 in the iterative solving process:
I z ＝max{[P-Q-G·(1-s)]/s, I z } (20)
F z ＝max{min[F z , 1], 0} (21)
through the process, the optimal reflectivity image F is finally obtained c (x) Noise Q in reflectance image c (x) Illumination image I c (x)。
And step S3: and correcting the reflectivity image and the illumination image through contrast stretching and brightness enhancement to obtain a restored underwater dark image.
Specifically, the luminance image I is corrected by using linear mapping c (x) And the reflectivity image F is improved by contrast stretching c (x) The contrast of (2); the linear mapping is shown in equations (22), (23), (24):
wherein the content of the first and second substances,
and &>
Are respectively an illuminance diagram I c (x) κ is a contrast adjustment parameter,>
and &>
Respectively representing images I c (x) Maximum and minimum values within the respective channels; />
Is the image after linear mapping and then is improved by gamma correction>
Is represented as:
wherein the content of the first and second substances,
and alpha is a brightness adjusting factor used for adjusting the brightness of the illumination map.
Increasing F by contrast stretch operator c (x) Is expressed as:
wherein the content of the first and second substances,
and &>
Respectively representing a maximum and a minimum of the reflectivity image, are present>
Namely the reflectivity image after contrast stretching.
Finally, the contrast stretched reflectivity image is processed
And the luminance corrected illuminance pattern>
Multiplying to obtain a restored underwater dark image:
where x is the pixel coordinate, c ∈ { r, g, b } is the red, green, blue three channels of the image, "" represents a point-to-point multiplication.
In the objective function formula (10), the linear mapping formula (24), and the gamma correction formula (25), there are five parameters φ 1 、φ 2 、φ 3 Kappa and alpha, the parameters in the existing underwater image restoration method are set according to experience and cannot be guaranteed to be optimal; this embodiment provides a static parameter optimization method for obtaining an optimal parameter phi 1 、φ 2 、v 3 Kappa and alpha, and the specific steps are as follows:
(1) Initialization: random initialization of a set y of 100 samples with a 16-bit binary code 1 ,y 2 ,…y i ,…y 100 Is denoted as Y u And the initial value of u is set to be 0, and the parameters to be optimized are 5: 1 、φ 2 、φ 3 κ, α, so the binary code for each sample is a 5 × 16 matrix, set Y u A three-dimensional matrix of 5 x 16 x 100.
(2) Calculating an image quality index of the sample: compute set Y one by one u The image quality indexes of all samples in the image processing system comprise a contrast index, a definition index and a color index, and specifically comprise the following steps:
inputting each sample into a formula (14), a formula (16), a formula (18), a formula (20), a formula (21) and a formula (27), and recovering 10 underwater dark images;
calculating image quality indexes of 10 restored images one by one;
the average of the quality indicators of the 10 images was taken as the final image quality indicator for this sample.
(3) Selection of an excellent sample: and screening excellent samples according to the image quality indexes of the samples. Firstly, according to the image quality index, calculating the normalized probability of 100 samples, then according to the normalized probability, executing 100 times of carousel selection, screening 100 samples to form a new set Y u+1 。
(4) Binary code exchange operation: randomly selecting a set Y u+1 Carrying out binary code exchange on the two samples in the system at random, judging whether the image quality index of the exchanged samples is larger than that of the samples before the exchange, and if so, not adopting the intersection result; the swap operation is performed 100 times in total, with the probability of performing the swap operation set to 50%.
(5) Mutation of binary code: for set Y u+1 The samples in the system are subjected to binary coding mutation operation one by one, namely one-bit binary codes of the samples are randomly selected to perform inversion operation; the probability of performing a mutation was set to 15%.
(6) And (4) terminating optimization: set Y u Obtaining a new set Y after selection operation, exchange operation and mutation operation u+1 (ii) a The above operation is executed circularly until u =15 stops; and finally, taking the sample with the maximum image quality index in the latest set as the optimal parameter.
In order to verify the effect of the embodiment, a plurality of underwater dark images are adopted for testing; as shown in fig. 2, the first behavior is an underwater dark image, the underwater dark image has low contrast and a dark color, and the second behavior is a restored image, which is restored by the method of the present embodiment, the image contrast is increased, the color is restored, the brightness is improved, and the noise is small; experimental results show that the method provided by the embodiment can obviously improve the visual effect of the underwater dark image.
Example two
The embodiment discloses an underwater dark image restoration system;
as shown in fig. 3, an underwater dark image restoration system includes a prior calculation module, a function solving module, and an image restoration module:
an a priori computation module configured to: calculating a background light value and a scene depth according to prior information of the underwater dark image;
a function solving module configured to: based on the background light value and the scene depth, combining an underwater scene decomposition model, constructing an objective function and solving, and calculating a reflectivity image, an illumination image and a noise item;
an image restoration module configured to: and correcting the reflectivity image and the illumination image through contrast stretching and brightness enhancement to obtain a restored underwater dark image.
EXAMPLE III
An object of the present embodiment is to provide a computer-readable storage medium.
A computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps in an underwater dark image restoration method according to one of the embodiments of the present disclosure.
Example four
An object of the present embodiment is to provide an electronic device.
The electronic device comprises a memory, a processor and a program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the steps in the underwater dark image restoration method according to the first embodiment of the disclosure.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An underwater dark image restoration method, comprising:
calculating a background light value and a scene depth according to prior information of the underwater dark image;
based on the background light value and the scene depth, combining an underwater scene decomposition model, constructing an objective function and solving, and calculating a reflectivity image, an illumination image and a noise item;
and correcting the reflectivity image and the illumination image through contrast stretching and brightness enhancement to obtain a restored underwater dark image.
2. The method for restoring the underwater dark image as claimed in claim 1, wherein the prior information of the underwater dark image comprises: and prior information among underwater dark image channels and prior information of image color and saturation.
3. The method for restoring the underwater dark image as claimed in claim 1, wherein the method for calculating the background light value comprises:
step (1): averagely dividing the underwater dark image into a plurality of regions, and calculating the score of each region by adopting an evaluation function;
step (2): selecting a region with the highest score as a candidate region, repeating the step (1) until the size of the candidate region is smaller than a preset threshold value, and outputting the region;
and (3): and calculating the pixel average value of the output area as the background light value.
4. The method for restoring the underwater dark image according to claim 1, wherein the underwater scene decomposition model specifically comprises:
P c (x)＝G c [1-s c (x)]+F c (x)·I c (x)·s c (x)+Q c (x)
wherein, the first and the second end of the pipe are connected with each other, c () Is a dark image under water and is a dark image, c is a value of the background light and is, c () As depth of scene, F c () As a reflectance image, I c () As an illumination image, Q c () Representing a noise item after illumination and scene depth attenuation, x being a pixel coordinate, c ∈ { r, } being a red channel, a green channel and a blue channel of an image, and' representing point-to-point multiplication.
5. The method for restoring the underwater dark image as claimed in claim 1, wherein the objective function is specifically:
wherein the content of the first and second substances,
holding a term for an unknown quantity>
In order to be a noise-suppressing term,
holding item, representing an illumination component>
Is a smooth term for the reflectivity>
Represents a gradient filtering template in a four-neighbor domain, | |. | non-woven phosphor 2 And | 1 Is L2 and L1 norm, respectively>
Representing convolution operations, I' representing a smooth underwater dark image, phi 1 、φ 2 、φ 3 Are three parameters that balance the weights of the terms in the objective function.
6. The underwater dark image restoration method according to claim 1, wherein the restored underwater dark image is specifically:
wherein, the first and the second end of the pipe are connected with each other,
is an illuminance diagram after correction of the brightness>
Is a reflectance image after contrast stretching, x is a pixel coordinate, c belongs to { r, g, b } is a red channel, a green channel and a blue channel of the image, and the ". The" represents point-to-point multiplication.
7. The method for restoring the underwater dark image according to claim 1, further comprising obtaining optimal parameters by a static parameter optimization method, specifically:
initializing a sample set;
calculating an image quality index of the sample;
screening out an excellent sample set according to the image quality indexes of all samples;
performing binary code exchange operation on the excellent sample set;
performing binary coding mutation operation on samples in the excellent sample set one by one;
and taking the sample with the largest image quality index in the excellent sample set as the optimal parameter.
8. An underwater dark image restoration system is characterized by comprising a priori calculation module, a function solving module and an image restoration module:
an a priori computation module configured to: calculating a background light value and a scene depth according to prior information of the underwater dark image;
a function solving module configured to: based on the background light value and the scene depth, combining an underwater scene decomposition model, constructing an objective function and solving, and calculating a reflectivity image, an illumination image and a noise item;
an image restoration module configured to: and correcting the reflectivity image and the illumination image through contrast stretching and brightness enhancement to obtain a restored underwater dark image.
9. An electronic device, comprising:
a memory for non-transitory storage of computer readable instructions; and
a processor for executing the computer readable instructions,
wherein the computer readable instructions, when executed by the processor, perform the method of any of claims 1-7.
10. A storage medium storing non-transitory computer-readable instructions, wherein the non-transitory computer-readable instructions, when executed by a computer, perform the instructions of the method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211632243.8A CN115937032A (en) | 2022-12-19 | 2022-12-19 | Underwater dark image restoration method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211632243.8A CN115937032A (en) | 2022-12-19 | 2022-12-19 | Underwater dark image restoration method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115937032A true CN115937032A (en) | 2023-04-07 |
Family
ID=86698951
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211632243.8A Pending CN115937032A (en) | 2022-12-19 | 2022-12-19 | Underwater dark image restoration method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115937032A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116777795A (en) * | 2023-08-21 | 2023-09-19 | 江苏游隼微电子有限公司 | Luminance mapping method suitable for vehicle-mounted image |
| CN117495845A (en) * | 2023-12-25 | 2024-02-02 | 宁德时代新能源科技股份有限公司 | Blue gel detection method and device, electronic equipment and medium |
-
2022
- 2022-12-19 CN CN202211632243.8A patent/CN115937032A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116777795A (en) * | 2023-08-21 | 2023-09-19 | 江苏游隼微电子有限公司 | Luminance mapping method suitable for vehicle-mounted image |
| CN116777795B (en) * | 2023-08-21 | 2023-10-27 | 江苏游隼微电子有限公司 | Luminance mapping method suitable for vehicle-mounted image |
| CN117495845A (en) * | 2023-12-25 | 2024-02-02 | 宁德时代新能源科技股份有限公司 | Blue gel detection method and device, electronic equipment and medium |
| CN117495845B (en) * | 2023-12-25 | 2024-05-10 | 宁德时代新能源科技股份有限公司 | Blue gel detection method and device, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112288658B (en) | Underwater image enhancement method based on multi-residual joint learning | |
| CN110232661B (en) | Low-illumination color image enhancement method based on Retinex and convolutional neural network | |
| CN115937032A (en) | Underwater dark image restoration method and system | |
| CN109410127B (en) | Image denoising method based on deep learning and multi-scale image enhancement | |
| CN111210395B (en) | Retinex underwater image enhancement method based on gray value mapping | |
| CN110288550B (en) | Single-image defogging method for generating countermeasure network based on priori knowledge guiding condition | |
| CN112381897B (en) | Low-illumination image enhancement method based on self-coding network structure | |
| CN111105371B (en) | Enhancement method of low-contrast infrared image | |
| CN109919859B (en) | Outdoor scene image defogging enhancement method, computing device and storage medium thereof | |
| CN111553863B (en) | Image enhancement method based on non-convex full-variation typing regularization | |
| CN110232668B (en) | Multi-scale image enhancement method | |
| CN113191983A (en) | Image denoising method and device based on deep learning attention mechanism | |
| CN114511480A (en) | Underwater image enhancement method based on fractional order convolution neural network | |
| CN113284061A (en) | Underwater image enhancement method based on gradient network | |
| CN115965544A (en) | Image enhancement method and system for self-adaptive brightness adjustment | |
| CN111563854B (en) | Particle swarm optimization method for underwater image enhancement processing | |
| KR102277005B1 (en) | Low-Light Image Processing Method and Device Using Unsupervised Learning | |
| CN117422653A (en) | Low-light image enhancement method based on weight sharing and iterative data optimization | |
| CN117392036A (en) | Low-light image enhancement method based on illumination amplitude | |
| CN116993616A (en) | Single low-illumination scene image enhancement method and enhancement system | |
| CN116862809A (en) | Image enhancement method under low exposure condition | |
| CN106127694B (en) | Adaptive two-way guarantor's bandwidth logarithmic transformation method of uneven illumination image enhancement | |
| CN110766616B (en) | Underwater image dodging algorithm based on single-scale Retinex method | |
| CN114529463A (en) | Image denoising method and system | |
| CN106952243A (en) | UUV Layer Near The Sea Surface infrared image self adaptation merger histogram stretches Enhancement Method |
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 |