CN105096272A - De-hazing method based on dual-tree complex wavelet - Google Patents
De-hazing method based on dual-tree complex wavelet Download PDFInfo
- Publication number
- CN105096272A CN105096272A CN201510512199.0A CN201510512199A CN105096272A CN 105096272 A CN105096272 A CN 105096272A CN 201510512199 A CN201510512199 A CN 201510512199A CN 105096272 A CN105096272 A CN 105096272A
- Authority
- CN
- China
- Prior art keywords
- image
- atmospheric
- dual
- dark
- tree complex
- 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
Abstract
The invention discloses a de-hazing method based on dual-tree complex wavelet and belongs to the technical field of image processing. The method comprises the following steps: to begin with, obtaining approximate value of atmospheric environment light through dual-tree complex wavelet transform; then, deducing a nonlinear normalized atmospheric transmissivity graph by utilizing an improved dark channel, and then, estimating a haze image; and finally, removing the haze to form a haze-free image according to the original image. The method has a better de-hazing effect for natural images of uniform haze and the like.
Description
Technical field
The present invention relates to image processing techniques, particularly a kind of defogging method based on dual-tree complex wavelet.
Background technology
Vision is one of the mankind main source obtaining external information, the approach obtained is also a lot, along with the development of modern science and technology, most visual information is nearly all by obtaining in image, owing to being subject to inclement weather (as mist, haze etc.) impact, picture contrast captured by imaging system obviously declines, details also can be fuzzy, and then real scene information at that time cannot be reflected, thus had a strong impact on military affairs, traffic, space flight and the analysis and decision further such as daily, therefore image demist has very important effect for outdoor study.
In recent years, image demist research also more and more receives the concern of lot of domestic and foreign researchist, propose many representational demist algorithms, these algorithms according to adopt image number be divided into roughly the demist based on multiple image and the demist based on single image.Based in the demist algorithm of several figure with Narasimhan propose algorithm for representative, utilize the light TRANSFER MODEL in atmospheric scattering to recover the image under mal-condition, reduce the concentration of mist in image, improve the sharpness (S.G.NarasimhanandS.K.Nayar of image, " VisionandtheAtmosphere; " InternationalJournalofComputerVision, 2002,48:233-254).The people such as Schechner eliminate in conjunction with the image of different polarization situation or reduce the impact of the mist in image, thus improve the display effect (TaliTreibitzandYoavY.Schechner of image, " Activepolarizationdescattering; " IEEETrans.PatternAnalysisandMachineIntelligence, 2009,31 (3): 385-399).The people such as domestic Chen Gong have first utilized mist and without the mist Computer image genration scene depth of field, then in conjunction with atmospherical scattering model remove mist (Chen Gong, Wang Tang, Zhou Heqin. the Misty Image of physically based deformation model restores new method. Journal of Image and Graphics.2008,13(5):888-893)。Different multiple images for Same Scene accurately obtain due to very difficult, and the demist algorithm range of application in actual scene based on multiple image is very narrow.And single image is very easy to obtain, increasing researcher turns to the demist of single image to study now.The people such as Tan propose the demist algorithm based on independent component analysis, this algorithm is found local contrast and is maximized the information improving image, thus the mist (R.Tan.Visibilityinbadweatherfromasingleimage.IEEEConfere nceonComputerVisionandPatternRecognition.2008,1-8) reduced in image.The people such as He are by studying a large amount of priori rules having found dark without mist image, combine monochromatic atmospherical scattering model simultaneously, effectively eliminate the mist (KaimingHe in image, JianSun, andXiaoouTang.Singleimagehazeremovalusingdarkchannelprio r.IEEEConferenceonComputerVisionandPatternRecognition.20 09,1956-1963).Fattal proposes the defogging method (R.Fattal.SingleImageDehazing.J.ACMSiggraph2008,1-9) based on new scene transmission.The people such as studies in China person Li Quan conjunction solve the unknown parameter in monochromatic atmospherical scattering model by new method, thus improve contrast and the sharpness (Li Quanhe of scene, Bi Duyan, Xu Yuelei, look into space. visible images scene reproduction .2014 under haze weather, 40 (4): 744-750).
The mist elimination algorithm of these physically based deformation models above-mentioned is nearly all by too much restrictive condition, solves the unknown parameter in relevant physical model, and then improves contrast or the sharpness of image.The value of these demist algorithms all assumptions' environment only same size, also only considers the single scale of image simultaneously, have ignored the impact of different scale information on image procossing.
At present, although the demist method comparison in some simple scenario is ripe, in the image that mist is more serious, removal effect is undesirable.If good demist model can be found, image mist elimination is very helpful, except, assessing easy degree, transplantability etc., is all the problem needing to consider.
Summary of the invention
For problems of the prior art, the invention provides a kind of approximate value utilizing even numbers complex wavelet transform to obtain the acquisition of information atmospheric environment light of image different scale, the first step first utilizes dual-tree complex wavelet transform to obtain the approximate value of surround lighting, then second step utilizes improvement dark channel prior knowledge to derive atmospheric transmissivity, then mist figure is estimated, last according to former figure, realize mist elimination, can solve in traffic, military, space flight, the erroneous judgement that the mist of numerous visual field such as remote sensing causes, improve obtain the quality of picture, strengthen the display effect of image, for further graphical analysis provides the defogging method of solid foundation.
Object of the present invention is achieved through the following technical solutions.
Based on a defogging method for dual-tree complex wavelet, step comprises:
1) original image pre-service: reduce image resolution ratio;
2) calculate atmospheric environment light: after image is carried out multilayer decomposition, extract the HFS of different directions, be then reconstructed, generate aerography being similar to as atmospheric environment light;
3) compute classes dark figure: theoretical based on dual-tree complex wavelet, image is first carried out multilayer decomposition, then every one deck is reconstructed from 6 directions, then theoretical according to dark, calculate the dark figure that every one deck is corresponding, finally using it on average as image class dark figure;
4) non-linear normalizing atmospheric transmissivity figure is calculated: according to step 3) the image class dark figure that generates calculates corresponding atmospheric transmissivity figure, and by its non-linear normalizing;
5) generate removal mist figure: according to atmospheric environment light and non-linear normalizing atmospheric transmissivity figure, calculate mist approximate diagram, then according to original image, generate mist elimination figure.
Described step 1) be specially: in RGB color image, according to the principle of arest neighbors, original image is reduced resolution to [480,640].
Described step 2) be specially: based on the HFS of multi-level Wavelet Transform, extract the information of all 1st layer, and merged formation approximate atmospheric surround lighting A
bwt.
Described step 3) in dual-tree complex wavelet theory be exactly on the basis of classical discrete wavelet theory, by 2 independently real small echo complete Phase information computing, i.e. Phase information
wherein,
it is all real small echo, every one deck Phase information decomposes the different directional information of rear 6 high frequency detail part correspondence image 6, namely ± 15 °, ± 45 °, then 6 of every one deck directional informations are reconstructed by the image information of ± 75 °, obtain the approximate value of each tomographic image, then according to the class dark figure of the every one deck of dark theory calculate, finally the class dark figure of all layers is averaged, as the class dark figure of original image.
Described step 4) in non-linear normalizing atmospheric transmissivity figure, first in step 3) basis on calculate atmospheric transmissivity figure: atmospheric transmissivity
Wherein, c represents the Color Channel of coloured image, and y is the neighborhood of pixel x, im
crepresent c passage original image,
be the class dark figure of c passage, owing to strengthening the effect of demist, undertaken non-linear by atmospheric transmissivity figure, process is as follows: map
natmo(x)=exp (map
atmo(x)/(2*sigma
2)), wherein, map
natmoit is the effect that nonlinear atmospheric transmissivity figure, sigma are used for controlling transmission.
Described step 5) be specially: according to approximate atmospheric surround lighting and non-linear normalizing atmospheric transmissivity figure, mist approximate diagram
wherein .* represents that approximate atmospheric surround lighting is multiplied with the pixel correspondence of non-linear normalizing, then according to original image, generates mist elimination figure.
Compared to prior art, the invention has the advantages that: the mist elimination algorithm of physically based deformation model is nearly all by restrictive condition, solves the unknown parameter in relevant physical model, thus improve contrast or the sharpness of image.But the value of the demist algorithm of these physically based deformation models all assumptions' environment only same size, also ignores image different scale information to the impact of image procossing simultaneously.The first step of the present invention first utilizes the low-frequency information of dual-tree complex wavelet transform analysis chart picture, obtain the approximate value of atmospheric environment light, then second step utilizes improvement dark channel prior knowledge to derive atmospheric transmissivity, then estimates mist figure, last according to former figure, realize mist elimination.Present invention utilizes the multi-scale information of dual-tree complex wavelet, improve mist elimination effect, calculate easy, realizability is good, therefore easily can be applied to computing machine, can be transplanted in embedded system after even optimizing.The present invention can be widely used in military affairs, traffic, space flight and remote sensing etc. outdoor study purposes.
Accompanying drawing explanation
Fig. 1 is the inventive method schematic flow sheet.
Fig. 2 is former figure and the atmospheric environment approximate diagram of the inventive method.
Fig. 3 is the former figure of the inventive method and mist layer approximate diagram.
Fig. 4 is the former figure of several special case of the inventive method and removes cloud and mist figure.
Embodiment
Below in conjunction with Figure of description and specific embodiment, the present invention is described in detail.
The present invention proposes based on two step defogging methods of dual-tree complex wavelet, the erroneous judgement that the mist for numerous visual field such as traffic, military affairs, space flight, remote sensing causes, and improves the quality that institute obtains picture, the display effect of enhancing image.The method has preferably except fog effect for the image under even haze, improves the contrast of image, enhances the display effect of image, is thus also a kind of effective image haze removal method.The first step of the present invention first utilizes the high-frequency information of dual-tree complex wavelet transform analysis chart picture, obtain the approximate value of atmospheric environment light, then second step utilizes improvement dark channel prior knowledge to derive atmospheric transmissivity, then estimates mist figure, last according to former figure, realize mist elimination.
Demist flow chart step of the present invention as shown in Figure 1, image is collected by collecting device, then by dual-tree complex wavelet level discharge rating, dual-tree complex wavelet transform is utilized to obtain the approximate value of atmospheric environment light, then second step utilizes improvement dark channel prior knowledge to derive atmospheric transmissivity, and then carries out non-linear normalizing, then estimates mist figure, last according to former figure, realize mist elimination.Specific as follows: a kind of defogging method based on dual-tree complex wavelet, step comprises:
1) original image pre-service: in order to improve the speed of calculating, reduces image resolution ratio, in RGB color image, according to the principle of arest neighbors, original image is reduced resolution to [480,640];
2) calculate atmospheric environment light: after image is carried out multilayer decomposition, extract the HFS of different directions, based on the HFS of multi-level Wavelet Transform, extract the information of all 1st layer, then be reconstructed, generate aerography being similar to as atmospheric environment light, be namely fused into approximate atmospheric surround lighting A
bwt.；
3) compute classes dark figure: theoretical based on dual-tree complex wavelet, by 2 independently real small echo complete Phase information computing, i.e. Phase information
wherein,
it is all real small echo, every one deck Phase information decomposes the different directional information of rear 6 high frequency detail part correspondence image 6, namely ± 15 °, ± 45 °, the image information of ± 75 °, more than classical wavelet transform 3 different directional informations, remain more image detail, then 6 of every one deck directional informations are reconstructed, obtain the approximate value of each tomographic image, then according to the dark theory (KaimingHe of the He Kaiming of Hong Kong Chinese University, JianSun, andXiaoouTang.Singleimagehazeremovalusingdarkchannelprio r.IEEETrans.PatternAnalysisandMachineIntelligence, 33 (12): 2341-2353) the class dark figure of every one deck is calculated, finally the class dark figure of all layers is averaged, as the class dark figure of original image.
4) calculate non-linear normalizing atmospheric transmissivity figure: first in step 3) basis on calculate atmospheric transmissivity figure: atmospheric transmissivity
Wherein, c represents the Color Channel of coloured image, and y is the neighborhood of pixel x, im
crepresent c passage original image,
be the class dark figure of c passage, owing to strengthening the effect of demist, undertaken non-linear by atmospheric transmissivity figure, process is as follows: map
natmo(x)=exp (map
atmo(x)/(2*sigma
2)), wherein, map
natmobe the effect that nonlinear atmospheric transmissivity figure, sigma are used for controlling transmission, its value is larger, and transmission is less, otherwise be worth less, transmission is larger.
5) removal mist figure is generated: according to approximate atmospheric surround lighting and non-linear normalizing atmospheric transmissivity figure, mist approximate diagram
Wherein .* represents that approximate atmospheric surround lighting is multiplied with the pixel correspondence of non-linear normalizing, then according to original image, generates mist elimination figure.
Fig. 2 is that dense fog is got down the highway former figure (a) and atmospheric environment propinquity effect figure (b).Can find that from former figure and atmospheric environment approximate diagram air approximate diagram is best being similar to of atmospheric environment at that time.
Fig. 3 is the approximate diagram (b) of former figure (a) and mist layer.
Mist figure under different scene and the design sketch after demist in the effect display set of Fig. 4, a () is the walkway behind walkway and mist elimination, b () is the toy after toy and mist elimination, c () is the highway after highway and mist elimination, the method that the above-mentioned the present invention of demonstrating proposes has preferably except fog effect for uniform mist.
Claims (6)
1., based on a defogging method for dual-tree complex wavelet, it is characterized in that step comprises:
1) original image pre-service: reduce image resolution ratio;
2) calculate atmospheric environment light: after image is carried out multilayer decomposition, extract the HFS of different directions, be then reconstructed, generate aerography being similar to as atmospheric environment light;
3) compute classes dark figure: theoretical based on dual-tree complex wavelet, image is first carried out multilayer decomposition, then every one deck is reconstructed from 6 directions, then theoretical according to dark, calculate the dark figure that every one deck is corresponding, finally using it on average as image class dark figure;
4) non-linear normalizing atmospheric transmissivity figure is calculated: according to step 3) the image class dark figure that generates calculates corresponding atmospheric transmissivity figure, and by its non-linear normalizing;
5) generate removal mist figure: according to atmospheric environment light and non-linear normalizing atmospheric transmissivity figure, calculate mist approximate diagram, then according to original image, generate mist elimination figure.
2. a kind of defogging method based on dual-tree complex wavelet according to claim 1, is characterized in that described step 1) be specially: in RGB color image, according to the principle of arest neighbors, original image is reduced resolution to [480,640].
3. a kind of defogging method based on dual-tree complex wavelet according to claim 1, is characterized in that described step 2) be specially: based on the HFS of multi-level Wavelet Transform, extract the information of all 1st layer, and merged formation approximate atmospheric surround lighting A
bwt.
4. a kind of defogging method based on dual-tree complex wavelet according to claim 1, it is characterized in that described step 3) in dual-tree complex wavelet theory be exactly on the basis of classical discrete wavelet theory, by 2 independently real small echo complete Phase information computing, i.e. Phase information
wherein,
it is all real small echo, every one deck Phase information decomposes the different directional information of rear 6 high frequency detail part correspondence image 6, namely ± 15 °, ± 45 °, then 6 of every one deck directional informations are reconstructed by the image information of ± 75 °, obtain the approximate value of each tomographic image, then according to the class dark figure of the every one deck of dark theory calculate, finally the class dark figure of all layers is averaged, as the class dark figure of original image.
5. a kind of defogging method based on dual-tree complex wavelet according to claim 1, is characterized in that described step 4) in non-linear normalizing atmospheric transmissivity figure, first in step 3) basis on calculate atmospheric transmissivity figure: atmospheric transmissivity
Wherein, c represents the Color Channel of coloured image, and y is the neighborhood of pixel x, im
crepresent c passage original image,
be the class dark figure of c passage, owing to strengthening the effect of demist, undertaken non-linear by atmospheric transmissivity figure, process is as follows: map
natmo(x)=exp (map
atmo(x)/(2*sigma
2)), wherein, map
natmoit is the effect that nonlinear atmospheric transmissivity figure, sigma are used for controlling transmission.
6. a kind of defogging method based on dual-tree complex wavelet according to claim 1, is characterized in that described step 5) be specially: according to approximate atmospheric surround lighting and non-linear normalizing atmospheric transmissivity figure, mist approximate diagram
wherein .* represents that approximate atmospheric surround lighting is multiplied with the pixel correspondence of non-linear normalizing, then according to original image, generates mist elimination figure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510512199.0A CN105096272B (en) | 2015-08-19 | 2015-08-19 | A kind of defogging method based on dual-tree complex wavelet |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510512199.0A CN105096272B (en) | 2015-08-19 | 2015-08-19 | A kind of defogging method based on dual-tree complex wavelet |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105096272A true CN105096272A (en) | 2015-11-25 |
| CN105096272B CN105096272B (en) | 2018-02-02 |
Family
ID=54576623
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510512199.0A Active CN105096272B (en) | 2015-08-19 | 2015-08-19 | A kind of defogging method based on dual-tree complex wavelet |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105096272B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107274369A (en) * | 2017-06-16 | 2017-10-20 | 南京信息职业技术学院 | A kind of single image defogging accelerated method based on frequency domain decomposition |
| CN107301633A (en) * | 2017-06-29 | 2017-10-27 | 常州工学院 | The emulation mode of remotely sensed image under cloud and mist interference |
| CN107977941A (en) * | 2017-12-04 | 2018-05-01 | 国网山东省电力公司电力科学研究院 | A kind of bright areas color fidelity and the image defogging method of contrast enhancing |
| CN108109129A (en) * | 2017-12-15 | 2018-06-01 | 四川大学 | A kind of rapid image defogging method based on near-infrared |
| CN109255748A (en) * | 2018-06-07 | 2019-01-22 | 上海出版印刷高等专科学校 | Digital watermark treatment method and system based on dual-tree complex wavelet |
| CN110717858A (en) * | 2019-10-09 | 2020-01-21 | 济源职业技术学院 | Image preprocessing method and device under low-illuminance environment |
| CN113379632A (en) * | 2021-06-15 | 2021-09-10 | 深圳市赛蓝科技有限公司 | Image defogging method and system based on wavelet transmissivity optimization |
| CN113406594A (en) * | 2021-06-01 | 2021-09-17 | 哈尔滨工业大学 | Single photon laser fog penetration method based on double-quantity estimation method |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000299781A (en) * | 1999-04-13 | 2000-10-24 | Matsushita Electric Ind Co Ltd | Image processing device and method |
| CA2507901A1 (en) * | 2004-05-21 | 2005-11-21 | Imaging Dynamics Company Ltd. | De-noising digital radiological images |
| CN101359399A (en) * | 2008-09-19 | 2009-02-04 | 常州工学院 | Cloud-removing method for optical image |
| CN102231791A (en) * | 2011-06-30 | 2011-11-02 | 北京云加速信息技术有限公司 | Video image defogging method based on image brightness stratification |
| CN102646267A (en) * | 2011-12-12 | 2012-08-22 | 中国科学院深圳先进技术研究院 | Degraded image restoration method and system |
| CN103955905A (en) * | 2014-05-13 | 2014-07-30 | 北京邮电大学 | Rapid wavelet transformation and weighted image fusion single-image defogging method |
| CN104200437A (en) * | 2014-09-04 | 2014-12-10 | 北京工业大学 | Image defogging method |
| CN104318528A (en) * | 2014-10-21 | 2015-01-28 | 北京工业大学 | Foggy weather image restoration method based on multi-scale WLS filtering |
| CN104537678A (en) * | 2015-01-15 | 2015-04-22 | 常州工学院 | Method for removing cloud and mist from single remote sensing image |
| CN104809709A (en) * | 2015-05-13 | 2015-07-29 | 北京邮电大学 | Single-image self-adaptation defogging method based on domain transformation and weighted quadtree decomposition |
-
2015
- 2015-08-19 CN CN201510512199.0A patent/CN105096272B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000299781A (en) * | 1999-04-13 | 2000-10-24 | Matsushita Electric Ind Co Ltd | Image processing device and method |
| CA2507901A1 (en) * | 2004-05-21 | 2005-11-21 | Imaging Dynamics Company Ltd. | De-noising digital radiological images |
| CN101359399A (en) * | 2008-09-19 | 2009-02-04 | 常州工学院 | Cloud-removing method for optical image |
| CN102231791A (en) * | 2011-06-30 | 2011-11-02 | 北京云加速信息技术有限公司 | Video image defogging method based on image brightness stratification |
| CN102646267A (en) * | 2011-12-12 | 2012-08-22 | 中国科学院深圳先进技术研究院 | Degraded image restoration method and system |
| CN103955905A (en) * | 2014-05-13 | 2014-07-30 | 北京邮电大学 | Rapid wavelet transformation and weighted image fusion single-image defogging method |
| CN104200437A (en) * | 2014-09-04 | 2014-12-10 | 北京工业大学 | Image defogging method |
| CN104318528A (en) * | 2014-10-21 | 2015-01-28 | 北京工业大学 | Foggy weather image restoration method based on multi-scale WLS filtering |
| CN104537678A (en) * | 2015-01-15 | 2015-04-22 | 常州工学院 | Method for removing cloud and mist from single remote sensing image |
| CN104809709A (en) * | 2015-05-13 | 2015-07-29 | 北京邮电大学 | Single-image self-adaptation defogging method based on domain transformation and weighted quadtree decomposition |
Non-Patent Citations (4)
| Title |
|---|
| ABHIJIT R. MANGE ET AL: ""Stabilizing Atmospheric Turbulence in a Video Surveillance using Dual Tree Complex Wavelet Fusion"", 《INTERNATIONAL JOURNAL OF COMPUTER APPLICATIONS (0975-8887)》 * |
| RAANAN FATTAL: ""Single Image Dehazing"", 《ACM TRANSACTIONS ON GRAPHICS》 * |
| 朱锡芳 等: ""基于Mallat算法遥感图像去云雾处理的改进方法"", 《遥感学报》 * |
| 李晓飞 等: ""基于父系数及邻域系数的双树复数小波变换算法在单幅彩色图像去雾中的应用"", 《电子技术》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107274369A (en) * | 2017-06-16 | 2017-10-20 | 南京信息职业技术学院 | A kind of single image defogging accelerated method based on frequency domain decomposition |
| CN107301633A (en) * | 2017-06-29 | 2017-10-27 | 常州工学院 | The emulation mode of remotely sensed image under cloud and mist interference |
| CN107301633B (en) * | 2017-06-29 | 2020-07-28 | 常州工学院 | Simulation method for remote sensing imaging under cloud and fog interference |
| CN107977941A (en) * | 2017-12-04 | 2018-05-01 | 国网山东省电力公司电力科学研究院 | A kind of bright areas color fidelity and the image defogging method of contrast enhancing |
| CN108109129A (en) * | 2017-12-15 | 2018-06-01 | 四川大学 | A kind of rapid image defogging method based on near-infrared |
| CN109255748A (en) * | 2018-06-07 | 2019-01-22 | 上海出版印刷高等专科学校 | Digital watermark treatment method and system based on dual-tree complex wavelet |
| CN110717858A (en) * | 2019-10-09 | 2020-01-21 | 济源职业技术学院 | Image preprocessing method and device under low-illuminance environment |
| CN113406594A (en) * | 2021-06-01 | 2021-09-17 | 哈尔滨工业大学 | Single photon laser fog penetration method based on double-quantity estimation method |
| CN113406594B (en) * | 2021-06-01 | 2023-06-27 | 哈尔滨工业大学 | Single photon laser fog penetrating method based on double-quantity estimation method |
| CN113379632A (en) * | 2021-06-15 | 2021-09-10 | 深圳市赛蓝科技有限公司 | Image defogging method and system based on wavelet transmissivity optimization |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105096272B (en) | 2018-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105096272A (en) | De-hazing method based on dual-tree complex wavelet | |
| Li et al. | Image dehazing using residual-based deep CNN | |
| Yang et al. | Low complexity underwater image enhancement based on dark channel prior | |
| CN102831591B (en) | Gaussian filter-based real-time defogging method for single image | |
| CN107103591B (en) | Single image defogging method based on image haze concentration estimation | |
| CN107301624B (en) | Convolutional neural network defogging method based on region division and dense fog pretreatment | |
| CN105096278B (en) | The image enchancing method adjusted based on illumination and equipment | |
| CN103049888A (en) | Image/video demisting method based on combination of dark primary color of atmospheric scattered light | |
| CN103985091A (en) | Single image defogging method based on luminance dark priori method and bilateral filtering | |
| CN109658447B (en) | Night image defogging method based on edge detail preservation | |
| WO2016169244A1 (en) | Method of denoising and enhancing video image based on random spray retinex and device utilizing same | |
| CN105931208A (en) | Physical model-based low-illuminance image enhancement algorithm | |
| CN102646267B (en) | Degraded image restoration method and system | |
| CN105957040A (en) | Rapid defog algorithm based on image fusion | |
| CN106530240A (en) | Image defogging method based on multi-scale fusion and total variational optimization | |
| Huang et al. | Towards unsupervised single image dehazing with deep learning | |
| Bansal et al. | A review of image restoration based image defogging algorithms | |
| Kumari et al. | Real time visibility enhancement for single image haze removal | |
| CN103778605A (en) | Greasy weather image enhancement method | |
| CN102968767A (en) | Method for real-time restoration of fog-degraded image with white balance correction | |
| Fu et al. | An anisotropic Gaussian filtering model for image de-hazing | |
| Li et al. | Research on haze image enhancement based on dark channel prior algorithm in machine vision | |
| Park et al. | Improved DCP-based image defogging using stereo images | |
| CN106780381B (en) | Low-illumination image self-adaptive enhancement method based on dark primary color and bilateral filtering | |
| Fang et al. | Detail Maintained Low-Light Video Image Enhancement Algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221214 Address after: No. 16, Gangxi Avenue, Menghe Town, Xinbei District, Changzhou City, Jiangsu Province, 213002 Patentee after: CHANGZHOU LEMENG PRESSURE VESSEL Co.,Ltd. Address before: 213022 Wushan Road, Xinbei District, Changzhou, Jiangsu Province, No. 1 Patentee before: CHANGZHOU INSTITUTE OF TECHNOLOGY |