CN113112429A - Universal enhancement framework for foggy images under complex illumination condition - Google Patents
Universal enhancement framework for foggy images under complex illumination condition Download PDFInfo
- Publication number
- CN113112429A CN113112429A CN202110457241.9A CN202110457241A CN113112429A CN 113112429 A CN113112429 A CN 113112429A CN 202110457241 A CN202110457241 A CN 202110457241A CN 113112429 A CN113112429 A CN 113112429A
- Authority
- CN
- China
- Prior art keywords
- layer
- detail
- base layer
- gamma
- estimation
- 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
- 238000005286 illumination Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000009826 distribution Methods 0.000 claims abstract description 11
- 230000005540 biological transmission Effects 0.000 claims abstract description 8
- 238000012937 correction Methods 0.000 claims abstract description 8
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 7
- 101100269850 Caenorhabditis elegans mask-1 gene Proteins 0.000 claims abstract description 6
- 239000000126 substance Substances 0.000 claims description 3
- 230000002708 enhancing effect Effects 0.000 claims description 2
- 238000002474 experimental method Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000011524 similarity measure Methods 0.000 description 2
- 230000003321 amplification Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 125000001475 halogen functional group Chemical group 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000002834 transmittance Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G06T5/70—
-
- G06T5/94—
Abstract
The invention discloses a universal enhancement framework for a foggy image under a complex illumination condition, which comprises the steps of constructing a decomposition model, and decomposing an input image into a basic layer and a detail layer; performing mask 1 estimation in the base layer to obtain the illumination distribution of the input image; the brightness of the base layer is improved, a gamma correction strategy is adopted to obtain a self-adaptive gamma factor, and the illumination of the base layer is adjusted according to the gamma factor to obtain the base layer with the improved brightness; performing mask 2 estimation on the detail layer to obtain refined detail mask estimation; carrying out transmission map estimation in an original image; designing a contrast enhancement factor; obtaining an enhanced detail layer according to the contrast enhancement factor; and reconstructing the basic layer with the improved brightness and the detail layer with the improved contrast to obtain the restored image. The method has higher universality and good robustness, can be suitable for low-light scenes with various illumination distributions, and obviously improves the visibility of underwater images.
Description
Technical Field
The invention relates to the field of images, in particular to a universal enhancement framework for fog images under complex illumination conditions.
Background
Many defogging methods have been proposed to improve the visibility of a blurred scene from a single image. Most pioneer works address this problem by inferring image degradation models, which is an inappropriate problem. Strong a priori knowledge or assumptions are required to estimate the parameters, especially the transmittance and the atmospheric light. Generally, the atmospheric light in a hazy image during the day is considered to be the brightest area in the background. However, the plurality of light sources seriously affect the accuracy of the estimation of the atmospheric light due to uneven illumination under night conditions. As a result, although methods based on Dark Channel Priors (DCP) 1 and variants [2] to [4] have been significantly successful, these demonstrate their effectiveness well only when demisting during the day, and are far from satisfactory in hazy scenes during the night.
To address this challenging problem, some researchers have focused on solving the nighttime defogging problem. Pei and Lee [5] first apply a color transfer function to reduce the color cast of the nightly hazy image before defogging based on a Dark Channel Prior (DCP). But due to the variation of the original color distribution, an artifact effect may occur. Similarly, Zhang [6] performs color correction and estimates the non-uniform incident illumination prior to DCP-based defogging. However, the light compensation will affect the effect of the subsequent color correction. Given that the local maximum intensity is mainly contributed by ambient lighting, Zhang [7] proposes a Maximum Reflectance Prior (MRP) that is specific to night time. Li 8 proposes a layer separation strategy to remove the halo of the light source by modeling the non-uniform illuminated night blurred image as a linear superposition of a low-light background and the lighting effect. Finally, a DCP-based defogging operation is followed to further improve visibility. As observed, the background recovered by these methods tends to be too dark. Unfortunately, all of the above methods are limited to handling nighttime conditions. To more accurately estimate atmospheric light and to be suitable for daytime and nighttime defogging, Anguti [9] introduced an efficient fusion-based technique that used local atmospheric light estimation. However, manually selecting the size of the patch is inefficient and the technique does not work well when lighting conditions are very poor. Zhu [10] introduces a Color Attenuation Prior (CAP), and analyzes a large amount of foggy images to find that the concentration of the fog changes along with the change of the depth of field, and the higher the concentration of the fog, the larger the depth of field, and the larger the difference between the brightness and the saturation of the images.
Reference documents:
[1]K.He,J.Sun,and X.Tang,“Single image haze removal using dark channel prior,”IEEE transactions on pattern analysis and machine intelligence,vol.33,no.12,pp.2341–2353,2010.
[2]H.Xu,J.Guo,Q.Liu,and L.Ye,“Fast image dehazing using improved dark channel prior,”in 2012IEEE International Conference on Information Science and Technology.IEEE,2012,pp.663–667.
[3]J.-B.Wang,N.He,L.-L.Zhang,and K.Lu,“Single image dehazing with a physical model and dark channel prior,”Neurocomputing,vol.149,pp.718–728,2015.
[4]Q.Wu,J.Zhang,W.Ren,W.Zuo,and X.Cao,“Accurate transmission estimation for removing haze and noise from a single image,”IEEE Transactions on Image Processing,vol.29,pp.2583–2597,2019.
[5]S.C.Pei and T.Y.Lee,“Nighttime haze removal using color transfer pre-processing and dark channel prior,”in Image Processing(ICIP),19th IEEE International Conference on,2012,pp.957–960.
[6]J.Zhang,Y.Cao,and Z.Wang,“Nighttime haze removal based on a new imaging model,”in IEEE International Conference on Image Processing(ICIP),2014,pp.4557–4561.
[7]J.Zhang,Y.Cao,S.Fang,Y.Kang,and C.W.Chen,“Fast haze removal for nighttime image using maximum reflectance prior,”in IEEE Conference on Computer Vision&Pattern Recognition(CVPR),2017,pp.7418–7426.
[8]Y.Li,R.T.Tan,and M.S.Brown,“Nighttime haze removal with glow and multiple light colors,”in IEEE International Conference on Computer Vision(ICCV),2015,pp.226–234.
[9]C.Ancuti,C.O.Ancuti,C.D.Vleeschouwer,and A.C.Bovik,“Day and night-time dehazing by local airlight estimation,”IEEE Transactions on Image Processing,vol.29,no.99,pp.6264–6275,2020.
[10]Q.Zhu,J.Mai,and L.Shao,“A fast single image haze removal algorithm using color attenuation prior,”IEEE transactions on image processing,vol.24,no.11,pp.3522–3533,2015.
disclosure of Invention
The present invention provides a generic enhancement framework for hazy images under complex lighting conditions to overcome the above technical problems.
A universal enhancement framework for fog images in complex lighting conditions, comprising,
the method comprises the following steps: constructing a decomposition model, and defining an input image by using a defogging model;
step two: a base-detail decomposition, which decomposes an original input image into a base layer and a detail layer;
step three: performing mask 1 estimation in the base layer, and comparing the average value of each channel in the base layer with the average value of the base layer to obtain the illumination distribution of the input image;
step four: the method comprises the steps of carrying out brightness improvement on a basic layer, obtaining a self-adaptive gamma factor by adopting a gamma correction strategy, and adjusting the illumination of the basic layer according to the gamma factor to obtain the basic layer with improved brightness;
step five: performing mask 2 estimation on the detail layer to obtain refined detail mask estimation;
step six: carrying out transmission map estimation in an original image;
step seven: estimating a contrast enhancement factor, and designing the contrast enhancement factor;
step eight: enhancing the detail layer, and obtaining the enhanced detail layer according to the contrast enhancement factor of the step seven;
step nine: and (5) reconstructing to obtain a restored image according to the base layer with the improved brightness obtained in the step four and the detail layer with the improved contrast obtained in the step eight.
Preferably, obtaining the base layer with improved brightness means performing correction through the formulas (1), (2) and (3),
gamma’＝||1-β*gamma|| (2)
where α and β are both constant parameters.
Preferably, designing the contrast enhancement factor means calculating by equation (4),
wherein the content of the first and second substances,
is a hyperbolic tangent function, gamma is a constant parameter controlling the sigmoid curve, and k is a contrast enhancement factor.
The universal enhancement framework for the foggy images under the complex illumination condition has higher universality, robustness and accuracy, and can be suitable for weak light scenes with various illumination distributions, such as non-uniform illumination images, pure low-illumination images, remote sensing images, underwater images and the like; layering the image to obtain two masks representing illumination distribution and effective detail area respectively, benefiting the two masks similar to the weight graph, and the proposed framework can achieve the effects of area self-adaptive illumination adjustment and contrast enhancement; qualitative and quantitative comparison shows that the method has excellent robustness, accuracy and flexibility.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a graph comparing PSNR evaluation results of the present invention;
FIG. 3 is a graph comparing SSIM evaluation results according to the present invention;
FIG. 4 is a graph of non-uniform illumination comparison results;
FIG. 5 is a graph of remote sensing image comparison results;
fig. 6 is a graph of underwater image comparison results.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flowchart of the present invention, and as shown in fig. 1, the method of this embodiment may include: a universal enhancement framework for hazy images in complex lighting conditions, comprising the steps of:
step 1: constructing a decomposition model, and defining an input image by using a defogging model;
where x is the pixel value, I (x) is the input image, J (x) is the enhanced image, t (x) is the transmission map, B is the global background light, l's(x) And l't(x) The enhanced base layer and detail layer are shown separately.
Step 2: the basic-detail decomposition is carried out,
considering the original input image as a superposition of two layers, namely a base layer and a detail layer, equation (1) can be simply expressed as:
wherein c belongs to { red, green, blue }, and represents three RGB color channels, Is(x) Is a base layer, It(x) Is a fine layer.
In this problem, a TV image reconstruction formula based on the rudin walker-Fatemi method is employed, and based on TV regularization, the base layer is obtained by minimizing the following objective function:
where x is the pixel index, λ is a parameter that adjusts the smoothness of the base layer,
is a gradient operator, which is a linear operator,
the gradient of the base layer is regularized by a 2 norm and the base layer is solved by TV-L2. Then according to the formula (3),
the fine layer I can be obtainedt(x)。
And step 3: the mask 1 estimation is performed in the base layer,
comparing the average value of each channel in the basic layer with the average value of the basic layer, namely when the average value of the three channels is larger than the average value of the basic layer, marking as 1, otherwise, marking as 0, and obtaining the illumination distribution M representing the input imageIIt can be expressed as:
wherein, in order to avoid artifacts, M isIObtaining refined M by using guided filteringI。
And 4, step 4: the brightness of the base layer is boosted by the brightness boosting circuit,
due to M in step 3IThe light source position can be indicated, therefore the invention proposes a new gamma correction strategy, namely, the difference value between the mask 1 and the average value thereof is utilized, when the pixel value of the mask 1 is larger than the average value thereof, the pixel value is larger, and the suppression is needed; on the contrary, the gamma factor can be obtained when enhancement is needed, and the self-adaptive gamma factor can be obtained by further constraining the gamma factor. The method can adaptively adjust the base layer I according to the distribution of lightsObtaining a luminance-enhanced base layer I'sThe formula can be expressed as:
gamma’＝||1-β*gamma|| (6)
wherein, both alpha and beta are constant parameters, beta is a constant for controlling the brightness improvement degree, and is set as 1.5 in the invention; and for IsIf the brightness is desired to be rapidly improved in the region with a small pixel value, otherwise, the brightness is improved in a smaller amplitude, and the gamma' is less than 1 according to the gamma correction principle; then 0 can be deduced<gamma<1.33; further push out 0.75<α<1 or 1<α<1.33. Selecting 10 underwater images with the numbers of a, b, c, d, f, g, h, i, j and k, and respectively taking alpha as [0.8,0.9,1.1,1.2,1.3 ]]And performing experimental verification, calculating the peak signal-to-noise ratio and the structural similarity measure of the image, and obtaining the peak signal-to-noise ratio (PSNR) and the structure of the enhanced image when alpha is 1.2The similarity measure (SSIM) works best, as shown in fig. 2, fig. 3; therefore, in the present invention, α is 1.2.
And 5: the mask 2 evaluation is performed in the detail layer,
using a Discrete Cosine Transform (DCT) on each 8 x 8 color block in the detail layer and representing the DCT coefficients of an 8 x 8 block as matrix a, the probability that the block is called part of the scene detail is expressed as:
where u, v represent positions in the DCT, and A is taken out1,1,A1,2,A2,1Except the sum of the squares of all DCT coefficients and thresholding the likelihood to binary divide each block.
Setting the threshold value to 0.1, i.e. when
And when the value is less than the threshold value, recording as 0, otherwise, recording as 1.
The initial detail mask estimate is denoted MTIt is refined using soft matting, i.e. the refined M is obtained by minimizing the following objective functionT：
argmin(mT-mT)T(mT-mT)+τmT TLsmT (9)
Wherein m isTAnd mTAre respectively a matrix
MTIs a regularization parameter, set to 10 in the present invention-5，LsIs a laplacian matrix generated from the original image i (x).
Step 6: the transmission map estimation is performed in the original image,
the atmospheric light B is estimated using the dark channel, first selecting the brightest 0.1% pixels in the dark channel, and then selecting the pixels of these pixels that result in the highest intensity in the image i (x) as atmospheric light.
First, converting I (x) -B into a space coordinate system, and then estimating an initial transmission diagram according to the value of the radius r (the distance from the origin, namely | | | I-B |)
Then through minimizing the pairs
To estimate the transmission map t (x):
where v, ω are weighting parameters, σ (x) is the standard deviation along each fog line, N (x) represents the four neighbors of pixel x, t (x) is the refined transmission map, t1(x)、
Are each t (x),
In the form of a vector.
And 7: contrast enhancement factor estimation
Having created mask 2 representing scene detail information, the present invention attempts to greatly enhance the effective detail without affecting the amplification of the fine detail (including noise). To achieve this goal, the present invention introduces a tan h function stretching MTTo scale the texture appropriately, the contrast enhancement factor can therefore be expressed as:
wherein the content of the first and second substances,
is a hyperbolic tangent function, and gamma is a constant controlling sigmoid curveThe parameter, set to 2.5 in the present invention.
And 8: the enhancement of the detail layer is carried out,
the enhanced detail layers can be obtained according to the contrast enhancement factor obtained in step 7 as follows:
I′t＝k·It (12)
and step 9: the reconstruction is carried out by the user,
according to the base layer with the improved brightness obtained in the step 4 and the detail layer with the improved contrast obtained in the step 8, the restored image can be obtained:
I′＝I′s+I′t (13)
six land images under non-uniform illumination conditions are selected, and comparison experiments are respectively carried out by six methods of DCP [1], CAP [10], Li et al [8], Zhang et al [6], MRP [7] and the invention, and the comparison experiment results are shown in figure 4. Two remote sensing images are selected, six methods including DCP [1], CAP [10], Li et al [8], Zhang et al [6], MRP [7] and the invention are respectively used for comparison experiments, and the comparison experiment results are shown in figure 5. Four underwater images are selected, and six methods including DCP [1], CAP [10], Li et al [8], Zhang et al [6], MRP [7] and the invention are respectively used for comparison experiments, and the comparison experiment results are shown in FIG. 6.
The beneficial effects of the whole are as follows:
(1) the proposed enhancement framework has higher universality, robustness and accuracy, and can be used for weak light scenes with various illumination distributions, such as non-uniform illumination images, pure low-illumination images, remote sensing images, underwater images and the like;
(2) layering the image to obtain two masks representing illumination distribution and effective detail area respectively, benefiting the two masks similar to the weight graph, and the proposed framework can achieve the effects of area self-adaptive illumination adjustment and contrast enhancement;
(3) qualitative and quantitative comparison shows that the method has excellent robustness, accuracy and flexibility.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (3)
1. A universal enhancement framework for fog images in complex lighting conditions, comprising,
the method comprises the following steps: constructing a decomposition model, and defining an input image by using a defogging model;
step two: a base-detail decomposition, which decomposes an original input image into a base layer and a detail layer;
step three: performing mask 1 estimation in the base layer, and comparing the average value of each channel in the base layer with the average value of the base layer to obtain the illumination distribution of the input image;
step four: the method comprises the steps of carrying out brightness improvement on a basic layer, obtaining a self-adaptive gamma factor by adopting a gamma correction strategy, and adjusting the illumination of the basic layer according to the gamma factor to obtain the basic layer with improved brightness;
step five: performing mask 2 estimation on the detail layer to obtain refined detail mask estimation;
step six: carrying out transmission map estimation in an original image;
step seven: estimating a contrast enhancement factor, and designing the contrast enhancement factor;
step eight: enhancing the detail layer, and obtaining the enhanced detail layer according to the contrast enhancement factor of the step seven;
step nine: and (5) reconstructing to obtain a restored image according to the base layer with the improved brightness obtained in the step four and the detail layer with the improved contrast obtained in the step eight.
2. The universal enhancement framework for fog images under complex lighting conditions as claimed in claim 1, wherein the base layer with enhanced brightness is obtained by correcting through equations (1), (2), (3),
gamma’＝||1-β*gamma|| (2)
where α and β are both constant parameters.
3. The universal enhancement framework for fog images under complex lighting conditions as claimed in claim 1, wherein the design contrast enhancement factor is calculated by formula (4),
wherein the content of the first and second substances,
is a hyperbolic tangent function, gamma is a constant parameter controlling the sigmoid curve, and k is a contrast enhancement factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110457241.9A CN113112429B (en) | 2021-04-27 | 2021-04-27 | Universal enhancement frame for foggy images under complex illumination conditions |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110457241.9A CN113112429B (en) | 2021-04-27 | 2021-04-27 | Universal enhancement frame for foggy images under complex illumination conditions |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113112429A true CN113112429A (en) | 2021-07-13 |
| CN113112429B CN113112429B (en) | 2024-04-16 |
Family
ID=76720145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110457241.9A Active CN113112429B (en) | 2021-04-27 | 2021-04-27 | Universal enhancement frame for foggy images under complex illumination conditions |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113112429B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5825936A (en) * | 1994-09-22 | 1998-10-20 | University Of South Florida | Image analyzing device using adaptive criteria |
| KR101448164B1 (en) * | 2013-04-22 | 2014-10-14 | 금오공과대학교 산학협력단 | Method for Image Haze Removal Using Parameter Optimization |
| WO2014193080A1 (en) * | 2013-05-28 | 2014-12-04 | 삼성테크윈 주식회사 | Method and device for removing haze in single image |
| WO2015103739A1 (en) * | 2014-01-08 | 2015-07-16 | 富士通株式会社 | Apparatus, electronic device and method for enhancing image contrast |
| US20160027160A1 (en) * | 2014-07-28 | 2016-01-28 | Disney Enterprises, Inc. | Temporally coherent local tone mapping of high dynamic range video |
| WO2016159884A1 (en) * | 2015-03-30 | 2016-10-06 | Agency For Science, Technology And Research | Method and device for image haze removal |
| CN106960421A (en) * | 2017-03-16 | 2017-07-18 | 天津大学 | Evening images defogging method based on statistical property and illumination estimate |
| CN108431886A (en) * | 2015-12-21 | 2018-08-21 | 皇家飞利浦有限公司 | Optimize high dynamic range images for particular display |
| CN110852956A (en) * | 2019-07-22 | 2020-02-28 | 江苏宇特光电科技股份有限公司 | Method for enhancing high dynamic range image |
| CN112634384A (en) * | 2020-12-23 | 2021-04-09 | 上海富瀚微电子股份有限公司 | Method and device for compressing high dynamic range image |
-
2021
- 2021-04-27 CN CN202110457241.9A patent/CN113112429B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5825936A (en) * | 1994-09-22 | 1998-10-20 | University Of South Florida | Image analyzing device using adaptive criteria |
| KR101448164B1 (en) * | 2013-04-22 | 2014-10-14 | 금오공과대학교 산학협력단 | Method for Image Haze Removal Using Parameter Optimization |
| WO2014193080A1 (en) * | 2013-05-28 | 2014-12-04 | 삼성테크윈 주식회사 | Method and device for removing haze in single image |
| WO2015103739A1 (en) * | 2014-01-08 | 2015-07-16 | 富士通株式会社 | Apparatus, electronic device and method for enhancing image contrast |
| US20160027160A1 (en) * | 2014-07-28 | 2016-01-28 | Disney Enterprises, Inc. | Temporally coherent local tone mapping of high dynamic range video |
| WO2016159884A1 (en) * | 2015-03-30 | 2016-10-06 | Agency For Science, Technology And Research | Method and device for image haze removal |
| CN108431886A (en) * | 2015-12-21 | 2018-08-21 | 皇家飞利浦有限公司 | Optimize high dynamic range images for particular display |
| CN106960421A (en) * | 2017-03-16 | 2017-07-18 | 天津大学 | Evening images defogging method based on statistical property and illumination estimate |
| CN110852956A (en) * | 2019-07-22 | 2020-02-28 | 江苏宇特光电科技股份有限公司 | Method for enhancing high dynamic range image |
| CN112634384A (en) * | 2020-12-23 | 2021-04-09 | 上海富瀚微电子股份有限公司 | Method and device for compressing high dynamic range image |
Non-Patent Citations (4)
| Title |
|---|
| EUNSUNG LEE ETAL.: "Contrast Enhancement Using Dominant Brightness Level Analysis and Adaptive Intensity Transformation for Remote Sensing Images", 《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》, vol. 10, no. 1, 31 December 2012 (2012-12-31), pages 62 - 66, XP055566774, DOI: 10.1109/LGRS.2012.2192412 * |
| 付青青;景春雷;裴彦良;阚光明;张正炳;吴爱平;: "基于非锐化掩模引导滤波的水下图像细节增强算法研究", 海洋学报, no. 07, 15 July 2020 (2020-07-15), pages 134 - 142 * |
| 杨爱萍;赵美琪;宋曹春洋;王金斌;: "基于色调映射和暗通道融合的弱光图像增强", 天津大学学报(自然科学与工程技术版), no. 07, 3 July 2018 (2018-07-03), pages 106 - 114 * |
| 田子建 等: "基于双域分解的矿井下图像增强算法", 《光子学报》, vol. 48, no. 5, 31 May 2019 (2019-05-31), pages 1 - 13 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113112429B (en) | 2024-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Enhancing underwater image via color correction and bi-interval contrast enhancement | |
| Bai et al. | Underwater image enhancement based on global and local equalization of histogram and dual-image multi-scale fusion | |
| Gao et al. | Sand-dust image restoration based on reversing the blue channel prior | |
| CN110782407B (en) | Single image defogging method based on sky region probability segmentation | |
| CN111080686B (en) | Method for highlight removal of image in natural scene | |
| Kumar et al. | An improved Gamma correction model for image dehazing in a multi-exposure fusion framework | |
| CN111861896A (en) | UUV-oriented underwater image color compensation and recovery method | |
| CN111968065A (en) | Self-adaptive enhancement method for image with uneven brightness | |
| Priyanka et al. | Low-light image enhancement by principal component analysis | |
| CN113284061A (en) | Underwater image enhancement method based on gradient network | |
| Zhu et al. | Underwater image enhancement based on colour correction and fusion | |
| CN114972102A (en) | Underwater image enhancement method based on global variable contrast enhancement and local correction | |
| CN116823686B (en) | Night infrared and visible light image fusion method based on image enhancement | |
| CN115619662A (en) | Image defogging method based on dark channel prior | |
| CN113112429B (en) | Universal enhancement frame for foggy images under complex illumination conditions | |
| CN113256533B (en) | Self-adaptive low-illumination image enhancement method and system based on MSRCR | |
| CN115908155A (en) | NSST domain combined GAN and scale correlation coefficient low-illumination image enhancement and denoising method | |
| CN115034985A (en) | Underwater image enhancement method | |
| CN114549342A (en) | Underwater image restoration method | |
| CN107317968A (en) | Image defogging method, device, computer can storage medium and mobile terminals | |
| Mi et al. | A generalized enhancement framework for hazy images with complex illumination | |
| CN106952243A (en) | UUV Layer Near The Sea Surface infrared image self adaptation merger histogram stretches Enhancement Method | |
| CN112529802B (en) | Atmospheric scattering degraded image recovery method based on scattering coefficient ratio estimation | |
| Deng | Mathematical approaches to digital color image denoising | |
| Radhika et al. | IMAGE CONTRAST ENCHANCEMENT USING BRIGHTNESS PRESERVATION DYNAMIC HISTOGRAM EQUALIZATION |
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 |