CN109242784A - A kind of haze weather atmosphere coverage rate prediction technique - Google Patents
A kind of haze weather atmosphere coverage rate prediction technique Download PDFInfo
- Publication number
- CN109242784A CN109242784A CN201810906689.2A CN201810906689A CN109242784A CN 109242784 A CN109242784 A CN 109242784A CN 201810906689 A CN201810906689 A CN 201810906689A CN 109242784 A CN109242784 A CN 109242784A
- Authority
- CN
- China
- Prior art keywords
- image
- atmosphere
- pixel
- haze weather
- coverage rate
- 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
- 238000000034 method Methods 0.000 title claims abstract description 19
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 8
- 238000011156 evaluation Methods 0.000 claims abstract description 6
- 238000009499 grossing Methods 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 abstract description 3
- 238000001914 filtration Methods 0.000 description 9
- 230000000007 visual effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000012887 quadratic function Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007850 degeneration Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of haze weather atmosphere coverage rate prediction techniques, it is characterized in that: covering parameter using Steerable filter algorithm evaluation current atmospheric.The present invention provides a kind of haze weather atmosphere coverage rate prediction techniques, it being capable of atmosphere coverage rate under Accurate Prediction haze weather, and then realization by the atmosphere covering analyzing to haze weather, and then realizes the image restoring to haze weather to the atmosphere coverage rate analysis of haze weather.
Description
Technical field
The present invention relates to computer disposal fields, more particularly to a kind of haze weather atmosphere coverage rate prediction technique.
Background technique
Computer vision system on present road is very sensitive to weather condition, wherein haze weather condition is various
To one kind of visual impact most serious in weather condition.Under the conditions of haze weather, visibility is greatly reduced, road environment system
The visual of system is deteriorated, and the image of acquisition receives serious degeneration, not only smudgy, and contrast reduces, but also can go out
Existing serious color displacement and distortion, many features contained in image are all capped, and what this greatly reduced image applies valence
Value, while also causing these vision systems that can not be worked normally.
In recent years, the sharpening processing of haze sky image has become the research of computer vision and field of image processing
Hot spot has attracted large quantities of researchers both domestic and external, and has had considerable method to be suggested.But due to weather condition itself
Complexity and randomness so that some algorithms proposed at present have certain limitation, some researchs obtained at
Fruit and research method wait further to improve still in continuous development.In the prior art, for road vision
System haze sky acquires the degenerate problem of image, it is not yet found that a kind of more effective haze weather image enchancing method, makes
It obtains under haze weather, can effectively realize image enhancement, increase the picture quality that effect improves vision signal, promote sensitivity.
And haze weather image enhancement is studied, the prediction of haze weather atmosphere coverage rate is most importantly carried out, still
In the prior art, less for the prediction of haze weather atmosphere coverage rate, lack a kind of more accurate haze weather atmosphere covering
Rate prediction technique.
Summary of the invention
In view of the above drawbacks of the prior art, technical problem to be solved by the invention is to provide a kind of more accurate
Haze weather atmosphere coverage rate prediction technique.
To achieve the above object, the present invention provides a kind of haze weather atmosphere coverage rate prediction techniques, it is characterized in that:
Parameter is covered using Steerable filter algorithm evaluation current atmospheric.
Preferably, it carries out calculating atmosphere covering parameter according to the following steps:
321) kernel of linear filter is established according to the following formula:
Wherein, wkIt is k-th of Kernel window；(ak,bk) it is the linear transform coefficient in a given cell window；
I is the pixel index in window；qiIt is output image；IiIt is the intensity of observed image；
322) nuance of the image p and output image q of input are distinguished according to the following formula:
Wherein, piIt is input picture；qiIt is output image；
323) solution of calculation formula (12) according to the following formula；
Wherein, brightness guiding image I is taken to be equal to input picture p；
At this moment, covk(I, p)=vark(I),We further obtain:
Wherein, ε is regularization smoothing factor；covk(I, p) is that the image of guiding figure I and input is covariance matrix；
vark(I) be guiding figure I variance matrix；It is the mean value of input picture；It is the mean value of guiding figure I；ε be it is smooth because
Son；
324) Steerable filter is applied to whole image region on the basis of retaining the hierarchical structure of original image, pressed
Atmosphere, which is calculated, according to following equation covers parameter:
I∞It is defined as
Wherein, | w | it is the pixel quantity in Kernel window；M is the index number of pixel；pmIt is the pixel of input picture；wm
It is and pmThe Kernel window of center pixel；Refer to the processing step of each pixel.
The beneficial effects of the present invention are: the present invention provides a kind of haze weather atmosphere coverage rate prediction technique, Neng Gouzhun
The really atmosphere coverage rate under prediction haze weather, and then realize the atmosphere coverage rate analysis to haze weather, by haze sky
The atmosphere covering analyzing of gas, and then realize the image restoring to haze weather.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of the embodiment of the invention.
Specific embodiment
Present invention will be further explained below with reference to the attached drawings and examples:
As shown in Figure 1, a kind of haze weather atmosphere coverage rate prediction technique, currently big using Steerable filter algorithm evaluation
Gas covers parameter.
Further, it carries out calculating atmosphere covering parameter according to the following steps:
321) kernel of linear filter is established according to the following formula:
Wherein, wkIt is k-th of Kernel window；(ak,bk) it is the linear transform coefficient in a given cell window；
I is the pixel index in window；qiIt is output image；IiIt is the intensity of observed image；
322) nuance of the image p and output image q of input are distinguished according to the following formula:
Wherein, piIt is input picture；qiIt is output image；
323) solution of calculation formula (12) according to the following formula；
Wherein, brightness guiding image I is taken to be equal to input picture p；
At this moment, covk(I, p)=vark(I),We further obtain:
Wherein, ε is regularization smoothing factor；covk(I, p) is that the image of guiding figure I and input is covariance matrix；
vark(I) be guiding figure I variance matrix；It is the mean value of input picture；It is the mean value of guiding figure I；ε be it is smooth because
Son；
324) Steerable filter is applied to whole image region on the basis of retaining the hierarchical structure of original image, pressed
Atmosphere, which is calculated, according to following equation covers parameter I∞:
I∞It is defined as
Wherein, | w | it is the pixel quantity in Kernel window；M is the index number of pixel；pmIt is the pixel of input picture；wm
It is and pmThe Kernel window of center pixel；Refer to the processing step of each pixel.
Using this method, the reinforcement of haze weather image, the increasing of especially suitable haze weather intelligent vehicular visual may be implemented
By force, a kind of haze weather intelligent vehicular visual Enhancement Method, comprising the following steps:
1) image is transformed into hsv color space from RGB color；
2) haze concentration is assessed；
3) it calculates atmosphere and covers parameter；
4) foreground image I is obtainedvis(x)；
5) current haze concentration is obtained as background image using the Steerable filter algorithm of luminance channel, utilize Gauss
Filter removes picture noise, increases influence of the atmosphere covering parameter to foreground image.
6) according to step 5), vision enhancement result is exported.
According to the dark priori theoretical of Dr.He show one clearly image in one of at least three RGB color channels
With low-pixel value, can indicate are as follows:
Jdark(x)=min (min (Jc(y))) c∈{r,g,b},y∈Ω(x) (9)
Here, JcIt is the Color Channel of RGB image J；Ω (x) is the eccentric statistical regions of x.
JdarkAlways very little, or even close to zero, in clearly image, in addition to sky area is largely clearly schemed
As statistical analysis.Image under low key tone image and different haze concentration is compared, by comparing low key tone part figure
Picture, it can be found that the mean pixel gray scale of low key tone image can be used for assessing the haze concentration of current environment.
Therefore, particularly, in the step 2), haze concentration is calculated according to the following steps:
21) evaluation points of haze concentration are defined according to the following formula:
Wherein, DdarkIt is dark primary priori；
U, v are the Fourier transformations of image pixel (x, y)；
vdark(x, y) is that one of at least three RGB color channels of pixel (x, y) have low-pixel value；
22) haze concentration is calculated according to the following formula:
fβ=k × Ddark
Wherein, k is linear scaling factor；fβIt is haze concentration,
During non-distorted image, fβThe limited range of parameter is limited to [2,8].
Particularly, in the step 3), parameter is covered using Steerable filter algorithm evaluation current atmospheric.
Particularly, it carries out calculating atmosphere covering parameter according to the following steps:
321) kernel of linear filter is established according to the following formula:
Wherein, wkIt is k-th of Kernel window；(ak,bk) it is the linear transform coefficient in a given cell window；
I is the pixel index in window；qiIt is output image；
322) nuance of the image p and output image q of input are distinguished according to the following formula:
Wherein, piIt is input picture；qiIt is output image；
323) solution of calculation formula (12) according to the following formula；
Wherein, brightness guiding image I is taken to be equal to input picture p；
At this moment, covk(I, p)=vark(I),We further obtain:
Wherein, ε is regularization smoothing factor；covk(I, p) is that the image of guiding figure I and input is covariance matrix；
vark(I) be guiding figure I variance matrix；It is the mean value of input picture；It is the mean value of guiding figure I；ε is smoothing factor.
324) Steerable filter is applied to whole image region on the basis of retaining the hierarchical structure of original image, pressed
Atmosphere, which is calculated, according to following equation covers parameter I∞:
I∞It is defined as
Wherein, | w | it is the pixel quantity in Kernel window；M is the index number of pixel；pmIt is the pixel of input picture；wm
It is and pmThe Kernel window of center pixel；Refer to the processing step of each pixel.In the present embodiment, smoothing factor ε is arranged
It is 0.3, in other embodiments, smoothing factor can also be configured according to different needs to reach identical technical effect,
The width r of Kernel window is set as 30 pixels.
Particularly, in the step 4), foreground image I is obtained according to the following formulavis(x):
Ivis(x)=E-I∞(18) wherein, E is the atmosphere light of infinite point
About the selection of filter, current most computer vision and computer graphics image filtering are related to
To the content for inhibiting or extracting image.Simply there is a constant filter of the linear translation of core (LTI), such as average, Gauss,
Laplce and Sobel filter are all widely used in image recovery, fuzzy/to sharpen, edge detection and feature extraction etc.
Deng.Selectable, LTI filter can be held explicitly by a Poisson's equation in high dynamic range (HDR) compression is solved
Row image mosaic, image scratch figure and gradient field operation, and filtering core is the transposition by a homogeneous Laplacian Matrix
Clearly it is defined.
LTI filtering core is space invariance and independent with picture material, but usually sometimes needs to consider guiding figure
The additional information of picture.The pioneering work of anisotropy parameter needs the gradient of the image filtered itself to go to instruct diffusion process,
It avoids smoothing to edge.Quadratic sum minimum weight filter needs the image filtered to go to instruct using input, and selects one
Quadratic function, this quadratic function are equivalent to the anisotropy parameter of a non-general stable state.In other applications,
Being oriented to image also can be other image, rather than original input picture.The output of Steerable filter device is the result is that be oriented to image
Local linear conversion.On the one hand, Steerable filter device has a good edge preserving smoothing effect, as two-sided filter, but
It is the influence that it does not have gradient to reverse artifact.On the other hand, Steerable filter can be far from smooth, in the auxiliary of guiding figure
It helps down, it can allow filtering to export more structuring, and smooth unlike input.
And in the technical program, the Steerable filter that we select mainly is accomplished by the following way:
We first define a common linear translation transformed filter program, with guiding image I, a filtering input
Image P and an output image q are related.I and p be according to application it is previously given, they can be identical.At one
Filter result at pixel i is to be expressed as a weighted average:
In formula, i and j are pixel subscripts.Filter kernel Wij is the function of guide image I, and independent with p.This
Filter is and p is linearly related.
The example of one such filter is that cascading filter volume bilateral filtering core Wbf is given by following formula
:
In formula, X is pixel coordinate；Ki is that a normalized parameter guarantees ∑ jWijbf=1；Parameter σsAnd σrIt adjusts separately
The sensitivity of space similarity and colour brightness range similarity.As P and I equal, cascading filter is degraded into initial pair
Side filter.
Explicit weighted average filter optimizes a quadratic function, and solves the linear equation of form below one
Group:
Aq=P (3)
Q and p is column vector N-1, corresponding { qi } and { pi }；A is the only relevant matrix with I of a N-N.Formula (3)
Q=A-1p is solved, has identical form, Wij=(A-1) ij with (1).
We define Steerable filter device now, and crucial hypothesis is this Steerable filter device in guiding image I and filters defeated
It is a Local Linear Model between q out.We assume that q be I center pixel k window Wk linear transformation:
(ak, bk) assumes that the identical linear coefficient of same Wk, the square window for being r with a radius.This local linear
As soon as long as model determination I has an edge, then q has an edge, because
In order to determine linear coefficient (ak, bk), the filtering image P for the input that needs restraint.We define output q as input p
Subtract some undesirable content n, such as noise/texture:
Qi=pi-ni (5)
We, which seek a solution, can minimize difference between q and p, while keep linear model (4).It is special
Other, we minimize the cost function of following window Wk:
Here, ∈ is the regularization parameter of the big ak of a punishment.
Equation (6) is linear ridge regression model, it is solved by formula (7) given below:
Here, μkWithIt is to be oriented to the window Wk average value of image I with variance；| W | it is number of pixels in Wk；
It is average value of the P in Wk.Linear coefficient (ak, bk) is obtained, we can calculate filtering
Export qi root.
However, a pixel i is related to all covering overlaid windows Wk of i, so the value of qi is not phase in equation (5)
Same working as is calculated in different windows.One simple strategy is the probable value for the qi being averaged out.So calculating institute
There is (ak, bk) value of Wk window in image, we, which calculate, exports result by following filtering:
Pay attention toSince box window is symmetrical, our rewrite equations (8) are as follows:
WithIt is that all overlaid windows at i are average
Coefficient.
The Average Strategy of this overlaid windows is popular in image denoising, and is very successfully.
Equation (6), (7), (8) are the definition of Steerable filter device.
Using this Steerable filter, the technical program simplifies atmospheric attenuation model.And based on dark-coloured priori theoretical assessment
Then haze concentration obtains current haze concentration as background image using the Steerable filter algorithm of luminance channel, increases big
Gas covers influence of the parameter to foreground image.Method provided by the invention can effectively improve the contrast and clarity of video,
Computational efficiency with higher simultaneously.
The preferred embodiment of the present invention has been described in detail above.It should be appreciated that those skilled in the art without
It needs creative work according to the present invention can conceive and makes many modifications and variations.Therefore, all technologies in the art
Personnel are available by logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Technical solution, all should be within the scope of protection determined by the claims.
Claims (2)
1. a kind of haze weather atmosphere coverage rate prediction technique, it is characterized in that: being covered using Steerable filter algorithm evaluation current atmospheric
Lid parameter.
2. haze weather atmosphere coverage rate prediction technique as described in claim 1, it is characterized in that:
It carries out calculating atmosphere covering parameter according to the following steps:
321) kernel of linear filter is established according to the following formula:
Wherein, wkIt is k-th of Kernel window；(ak,bk) it is the linear transform coefficient in a given cell window；I be
Pixel index in window；qiIt is output image；IiIt is the intensity of observed image；
322) nuance of the image p and output image q of input are distinguished according to the following formula:
Wherein, piIt is input picture；qiIt is output image；
323) solution of calculation formula (12) according to the following formula；
Wherein, brightness guiding image I is taken to be equal to input picture p；
At this moment, covk(I, p)=vark(I),We further obtain:
Wherein, ε is regularization smoothing factor；covk(I, p) is that the image of guiding figure I and input is covariance matrix；vark(I)
It is the variance matrix of guiding figure I；It is the mean value of input picture；It is the mean value of guiding figure I；ε is smoothing factor；
324) Steerable filter is applied to whole image region on the basis of retaining the hierarchical structure of original image, according to following
Formula calculates atmosphere and covers parameter I∞:
I∞It is defined as
Wherein, | w | it is the pixel quantity in Kernel window；M is the index number of pixel；pmIt is the pixel of input picture；wmIt is and pm
The Kernel window of center pixel；Refer to the processing step of each pixel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810906689.2A CN109242784A (en) | 2018-08-10 | 2018-08-10 | A kind of haze weather atmosphere coverage rate prediction technique |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810906689.2A CN109242784A (en) | 2018-08-10 | 2018-08-10 | A kind of haze weather atmosphere coverage rate prediction technique |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109242784A true CN109242784A (en) | 2019-01-18 |
Family
ID=65071168
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810906689.2A Pending CN109242784A (en) | 2018-08-10 | 2018-08-10 | A kind of haze weather atmosphere coverage rate prediction technique |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109242784A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104683767A (en) * | 2015-02-10 | 2015-06-03 | 浙江宇视科技有限公司 | Fog penetrating image generation method and device |
| US9349170B1 (en) * | 2014-09-04 | 2016-05-24 | The United States Of America As Represented By The Secretary Of The Navy | Single image contrast enhancement method using the adaptive wiener filter |
| CN105827976A (en) * | 2016-04-26 | 2016-08-03 | 北京博瑞空间科技发展有限公司 | GPU (graphics processing unit)-based video acquisition and processing device and system |
| CN106127706A (en) * | 2016-06-20 | 2016-11-16 | 华南理工大学 | A kind of single image defogging method based on non-linear cluster |
| CN107133927A (en) * | 2017-04-21 | 2017-09-05 | 汪云飞 | Single image to the fog method based on average mean square deviation dark under super-pixel framework |
-
2018
- 2018-08-10 CN CN201810906689.2A patent/CN109242784A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9349170B1 (en) * | 2014-09-04 | 2016-05-24 | The United States Of America As Represented By The Secretary Of The Navy | Single image contrast enhancement method using the adaptive wiener filter |
| CN104683767A (en) * | 2015-02-10 | 2015-06-03 | 浙江宇视科技有限公司 | Fog penetrating image generation method and device |
| CN105827976A (en) * | 2016-04-26 | 2016-08-03 | 北京博瑞空间科技发展有限公司 | GPU (graphics processing unit)-based video acquisition and processing device and system |
| CN106127706A (en) * | 2016-06-20 | 2016-11-16 | 华南理工大学 | A kind of single image defogging method based on non-linear cluster |
| CN107133927A (en) * | 2017-04-21 | 2017-09-05 | 汪云飞 | Single image to the fog method based on average mean square deviation dark under super-pixel framework |
Non-Patent Citations (1)
| Title |
|---|
| 夏玉琪等: "基于透射率估计模型和引导滤波的雾气图像去雾算法", 《西华大学学报(自然科学版)》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Wavelet-based dual-branch network for image demoiréing | |
| Singh et al. | A comprehensive review of computational dehazing techniques | |
| CN109214993A (en) | A kind of haze weather intelligent vehicular visual Enhancement Method | |
| Park et al. | Single image dehazing with image entropy and information fidelity | |
| Singh et al. | Dehazing of outdoor images using notch based integral guided filter | |
| Ma et al. | An effective fusion defogging approach for single sea fog image | |
| Li et al. | Spectrum characteristics preserved visible and near-infrared image fusion algorithm | |
| Zhu et al. | Multiscale infrared and visible image fusion using gradient domain guided image filtering | |
| CN108765336A (en) | Image defogging method based on dark bright primary colors priori with auto-adaptive parameter optimization | |
| CN110097522A (en) | A kind of single width Method of defogging image of outdoor scenes based on multiple dimensioned convolutional neural networks | |
| CN112991197B (en) | Low-illumination video enhancement method and device based on detail preservation of dark channel | |
| CN111754433B (en) | Defogging method for aerial image | |
| Zou et al. | Infrared and low-light visible image fusion based on hybrid multiscale decomposition and adaptive light adjustment | |
| Liu et al. | Single image haze removal via depth-based contrast stretching transform | |
| Shi et al. | A joint deep neural networks-based method for single nighttime rainy image enhancement | |
| CN115100226A (en) | Contour extraction method based on monocular digital image | |
| Fuh et al. | Mcpa: A fast single image haze removal method based on the minimum channel and patchless approach | |
| CN109255804A (en) | A kind of haze concentration sealing method | |
| Li et al. | DLT-Net: deep learning transmittance network for single image haze removal | |
| Singh et al. | Multiscale reflection component based weakly illuminated nighttime image enhancement | |
| CN109242784A (en) | A kind of haze weather atmosphere coverage rate prediction technique | |
| Deshpande et al. | An efficient image deblurring method with a deep convolutional neural network for satellite imagery | |
| CN110633705A (en) | Low-illumination imaging license plate recognition method and device | |
| Sun et al. | Single-image dehazing based on dark channel prior and fast weighted guided filtering | |
| Ni et al. | A fast sand-dust video quality improvement method based on adaptive dynamic guided filtering and interframe detection strategy |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190118 |