CN112927157B - Improved dark channel defogging method adopting weighted least square filtering - Google Patents
Improved dark channel defogging method adopting weighted least square filtering Download PDFInfo
- Publication number
- CN112927157B CN112927157B CN202110249791.1A CN202110249791A CN112927157B CN 112927157 B CN112927157 B CN 112927157B CN 202110249791 A CN202110249791 A CN 202110249791A CN 112927157 B CN112927157 B CN 112927157B
- Authority
- CN
- China
- Prior art keywords
- image
- transmission rate
- dark channel
- matrix
- defogging
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000001914 filtration Methods 0.000 title claims abstract description 16
- 230000005540 biological transmission Effects 0.000 claims abstract description 40
- 239000011159 matrix material Substances 0.000 claims description 20
- 238000010586 diagram Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000009499 grossing Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 230000017105 transposition Effects 0.000 claims description 2
- 125000001475 halogen functional group Chemical group 0.000 description 3
- 235000000832 Ayote Nutrition 0.000 description 2
- 241000219122 Cucurbita Species 0.000 description 2
- 235000009854 Cucurbita moschata Nutrition 0.000 description 2
- 235000009804 Cucurbita pepo subsp pepo Nutrition 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 235000015136 pumpkin Nutrition 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000013618 particulate matter Substances 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
Classifications
-
- G06T5/73—
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses an improved dark channel defogging method adopting weighted least square filtering, which is applied to the field of image processing, and aims at the phenomenon that a dark-pass algorithm adopting guided filtering to refine a transmission rate graph in the existing dark channel defogging algorithm can generate more obvious halation after defogging; so as to eliminate obvious halation phenomenon on the premise of ensuring that defogging quality is basically unchanged.
Description
Technical Field
The invention belongs to the field of image processing, and particularly relates to an image defogging technology.
Background
Images acquired outdoors often suffer from poor image quality due to factors such as the effects of various particulate matter in the atmosphere due to non-ideal weather conditions. Such images are unsatisfactory in terms of both photographic imaging and quality as input to subsequent higher order processing. Along with the upgrading and penetration of the digital industry, the image processing and computer vision technology is widely applied in production and life, and simultaneously, higher requirements are also put on the imaging quality. For example, monitoring and autopilot systems, images acquired under the influence of foggy weather are often accompanied by problems such as reduced contrast, color distortion, and occlusion of features. In this situation the performance of the system is necessarily lost and even not working properly. The defogging algorithm is introduced, so that the influence of haze can be eliminated to a certain extent, the original details of the image are displayed, the image is clearer, and meanwhile, the distortion in brightness, contrast and color is corrected.
In recent years, a number of methods have been proposed for defogging of single images, wherein the prior defogging of a dark channel of He Kaiming is used as a classical traditional defogging algorithm, and is always a marker post compared with subsequent workers. But the dark-pass algorithm for refining the transmission rate diagram by adopting the guide filtering can generate a more obvious halation phenomenon after defogging.
Disclosure of Invention
In order to solve the technical problems, the invention provides an improved dark channel defogging method adopting weighted least square filtering, which eliminates obvious halation.
The invention adopts the technical scheme that: the improved dark channel defogging method adopting the weighted least square filter is used for correcting the transmission rate diagram by adopting the weighted least square filter.
A1, estimating an atmospheric light value according to the acquired foggy image;
a2, carrying out normalization processing on the acquired foggy image by utilizing an atmospheric light value;
a3, obtaining a dark channel image of a single pixel of the image normalized by the step A2x is the position of the pixel point;
a4, acquiring a dark channel image of the image normalized by the step A2 with a set radius r;
a5, calculating a transmission rate reliability weight delta according to the results obtained in the step A3 and the step A4;
a6, calculating a transmission rate estimation graph according to the atmospheric light value and the result of the step A4;
a7, carrying out weighted least square filtering processing on the transmission rate estimation graph according to the transmission rate credibility weight value;
and A8, defogging the acquired fogged image by using the atmospheric light value and the transmission rate image processed in the step A7.
The specific process of the step A7 is as follows:
the following cost function is minimized:
wherein ,for matrix->Element of (a)>For matrix->Element N of (3) x A four-neighborhood representing a position x, y representing N x Coordinates of (E) above>I (y) represents N x Corresponding values are added;
the solution is as follows:
wherein ,for the corrected transmission rate estimation graph, delta represents the transmission rate reliability weight value obtained in the step A5, and ++>Representing the transmission rate estimation graph obtained in the step A6, D x 、D y Respectively with fog images IThe difference in the x, y directions, λ is the smoothing coefficient, and T in the superscript stands for transpose.
The specific process for solving the cost function is as follows:
a71, changing the foggy image into a gray level image;
a72, calculating the affinity between adjacent pixels according to the gradient of the gray level map;
D x ＝-λ./[G(x+1,y)-G(x,y)]
D y ＝-λ./[G(x,y+1)-G(x,y)]
D x ，D y the affinities of adjacent pixels in the x, y directions, respectively, G (x, y) representing the gray value at (x, y);
a73, constructing a Laplacian homogeneous matrix by using the result in the step A72;
a74, adding the main diagonal elements of the Laplacian homogeneous matrix obtained in the step A73 with the elements in the weight delta respectively to obtain a parameter matrix M;
a75, element in weight delta and transmission rate estimation graphSequentially multiplying and expanding the elements in the sequence into a column vector N;
wherein N (x) represents an x-th element in N;
a76, multiplying the inverse point of the matrix M obtained in the step A74 by the column vector N obtained in the step A75;
V＝M -1 ·N
a77, rearranging the result V of the step A76 into a matrix with the same height and width as the image to obtain a corrected transmission rate diagram
The calculation formula of the step A8 is as follows:
wherein J (x) represents an image after defogging.
The calculation formula of the step A6 is as follows:
wherein, delta is the weight,H max is the maximum value in matrix H.
The calculation formula of the step A6 is as follows:
wherein, omega (x) is a window taking x as a center r as a radius, a is the number of pixel points contained in the omega (x) window,and (5) obtaining a single-pixel dark channel diagram in the step four.
The invention has the beneficial effects that: in defogging algorithms, the magnitude of the transmission rate is directly related to the distance of the scene from the camera sensor. Dark channel algorithm in estimating the transmission rate map, if objects with larger distance are included in one window at the same time, the value of the pixel point with the closer distance is usually selected as the dark channel. The transmission values obtained in this case are inaccurate, so that halos remain on the edges of some objects after defogging; according to the invention, the transmission rate diagram is corrected by changing the guide filtering in the original algorithm into the weighted least square filtering, so that punishment on an unreliable estimated value can be introduced, and a larger transmission rate estimated value can be obtained by inhibiting the distant view. So as to eliminate obvious halation phenomenon on the premise of ensuring that defogging quality is basically unchanged.
Drawings
FIG. 1 is a flow chart of a defogging method according to the present invention;
fig. 2 is a defogging contrast of a foggy image pumpkins;
wherein, fig. 2 (a) is an original chart, fig. 2 (b) is a He algorithm processing result, and fig. 2 (c) is a processing result of the present invention;
FIG. 3 is a defogging contrast of the foggy image train;
wherein, fig. 3 (a) is an original chart, fig. 3 (b) is a He algorithm processing result, and fig. 3 (c) is a processing result of the present invention;
FIG. 4 is a defogging contrast of a foggy image cityscape;
wherein, fig. 4 (a) is an original chart, fig. 4 (b) is a He algorithm processing result, and fig. 4 (c) is a processing result of the present invention;
FIG. 5 is a defogging contrast of a foggy image canyon;
wherein, fig. 5 (a) is an original chart, fig. 5 (b) is a He algorithm processing result, and fig. 5 (c) is a processing result of the present invention;
FIG. 6 is a defogging contrast of the foggy image forest;
wherein, fig. 6 (a) is an original image, fig. 6 (b) is a He algorithm processing result, and fig. 6 (c) is a processing result of the present invention;
FIG. 7 is a plot of the transmission rate of an algorithmically processed fog image forest of He;
fig. 8 is a transmission rate graph of a fog image forest processed by the method of the present invention.
Detailed Description
The present invention will be further explained below with reference to the drawings in order to facilitate understanding of technical contents of the present invention to those skilled in the art.
As shown in fig. 1, an improved dark channel defogging method using weighted least squares filtering of the present invention comprises the steps of:
s1, acquiring foggy image data I;
s2, estimating an atmospheric light value; the first 0.1% pixel is taken from the dark channel map as the luminance size. In these positions, the value corresponding to the point with the highest brightness is found in the original image as a value of a, which represents the atmospheric light value.
S3, normalizing the fogged image by using an atmospheric light value; the specific process is as follows:
wherein P is a normalized image, A is an atmospheric light value, and I is a fogged image;
s4, obtaining a dark channel image of a single pixel of the image P; the calculation formula is as follows:
wherein x is the position of the pixel point,for obtaining a single-pixel dark channel image, c refers to three channels R, G and B and P c I.e. the pixel values in the three channels R, G, B respectively.
S5, acquiring a dark channel image of the image P by r; the specific process is as follows:
wherein ,for manipulating dark channel images with a window radius r, < >>Omega (x) is a window with x as a center r as a radius for the single pixel dark channel value obtained in the previous step; the window radius r is generally 15, and in practice, the window radius r is 5 or more.
S6, calculating a transmission rate reliability weight value according to the result of the step S4 and the step S5; the specific process is as follows:
wherein, delta is the weight,H max is the maximum value in matrix H; the weight calculation mode provided by the method is easy to realize, and the calculation cost can be greatly reduced.
S7, according to the dark channel diagramCalculating a transmission rate estimation graph with the atmospheric light value A; the specific process is as follows:
wherein For the transmission rate estimation graph, ω is the defogging index, with a typical value of 0.95;
s8, carrying out weighted least square filtering processing on the transmission rate estimation graph; the specific process is as follows:
minimizing cost function:
the solution is as follows:
wherein ,is corrected byTransmission rate diagram, D x ，D y Taking 0.1 for the difference of the foggy image I in the x and y directions, wherein lambda is a smoothing coefficient, and T in the superscript represents transposition; the specific process of calculating the solution is as follows:
s81, changing the foggy image into a gray level image;
s82, calculating the affinity between adjacent pixels according to the gradient of the gray level map;
D x ＝-λ./[G(x+1,y)-G(x,y)]
D y ＝-λ./[G(x,y+1)-G(x,y)]
D x ，D y the affinities of adjacent pixels in the x, y directions, respectively, G (x, y) represents the gray value at (x, y).
S83, constructing a Laplacian homogeneous matrix by using the result in the step S82;
s84, adding the main diagonal elements of the Laplacian matrix obtained in the step S83 with the elements in the weight delta respectively, and finally obtaining a parameter matrix M in the form as follows:
wherein ,Dy1 、D x1 、D y2 、D x2 、...、D yh 、D xh Is an element in the Laplacian matrix;
s85, element in weight delta and transmission rate estimation diagramSequentially multiplying and expanding the elements in the sequence into a column vector N;
s86, multiplying the inverse of the matrix M obtained in the step S84 by the column vector N obtained in the step S85;
V＝M -1 ·N
s87, rearranging the result V of the step S86 into a moment with the same height and width as the imageThe matrix is used for obtaining the corrected transmission rate diagram
S9, defogging the fogged image by using the obtained atmospheric light value and the corrected transmission rate graph; the process comprises the following steps:
wherein J is the finally obtained defogged image.
The invention also provides another mode for calculating the reliability weight value of the transmission rate, and the specific calculation formula is as follows:
wherein, omega (x) is a window taking x as a center r as a radius, a is the number of pixel points contained in the omega (x) window,and (5) obtaining a single-pixel dark channel diagram in the step four. The weight calculation method provided in this embodiment can better embody the transmission rate map +.>The results obtained theoretically would be better.
Fig. 2, 3, 4, 5, 6 are a comparison of the algorithm of using He for different hazed images (i.e., the dark channel prior defogging of He Kaiming as classical conventional defogging algorithm) with the method of the present invention. As shown in fig. 2, the defogging contrast of the foggy image pumpkins is shown in fig. 2 (a) as original image, fig. 2 (b) as He algorithm processing result, and fig. 2 (c) as processing result of the present invention; as shown in fig. 3, the defogging contrast of the foggy image train is shown in fig. 3 (a) as original image, fig. 3 (b) as He algorithm processing result, and fig. 3 (c) as processing result of the present invention; in such a continuous scenario as in fig. 2 and 3, the algorithm of He is almost identical to the method of the present invention in terms of defogging results.
As shown in fig. 4, the defogging contrast of the foggy image cityscape is shown in fig. 4 (a) as original image, fig. 4 (b) as He algorithm processing result, and fig. 4 (c) as processing result of the present invention; as shown in fig. 5, the defogging contrast of the foggy image canyon is shown in fig. 5 (a) as original image, fig. 5 (b) as He algorithm processing result, and fig. 5 (c) as processing result of the present invention; however, when the image includes a near view and a far view with a large difference in distance as in fig. 4 and 5, the halation phenomenon is liable to occur under the He algorithm. As in the junction of the near tree and the far road in FIG. 4, the algorithm of He has obvious white halo, and the surrounding of the left and right trees in FIG. 5 has similar phenomena. The method of the present invention can effectively suppress this phenomenon in the scenario of fig. 4 and 5.
As shown in fig. 6, the defogging contrast of the foggy image forest is shown in fig. 6 (a) as original image, fig. 6 (b) as He algorithm processing result, and fig. 6 (c) as processing result of the present invention; the middle dense fog portion of fig. 6 (a) is processed into a far scene in the He algorithm (see fig. 7), thus employing more aggressive defogging, resulting in halos around the branches and leaves near the middle. The method of the present invention can suppress the edge halation on the one hand and alleviate the halation on the other hand when dealing with the scene (see fig. 8) where the thick fog portion is not regarded as sky as far as the transmission rate map is corrected.
Those of ordinary skill in the art will recognize that the embodiments described herein are for the purpose of aiding the reader in understanding the principles of the present invention and should be understood that the scope of the invention is not limited to such specific statements and embodiments. Various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (4)
1. An improved dark channel defogging method using weighted least squares filtering, comprising:
a1, estimating an atmospheric light value according to the acquired foggy image;
a2, carrying out normalization processing on the acquired foggy image by utilizing an atmospheric light value;
a3, obtaining a dark channel image of a single pixel of the image normalized by the step A2x is the position of the pixel point;
a4, acquiring a dark channel image of the image normalized by the step A2 with a set radius r; the specific process is as follows:
wherein ,for manipulating dark channel images with a window radius r, < >>Omega (x) is a window with x as a center r as a radius for the single pixel dark channel value obtained in the previous step;
a5, calculating a transmission rate reliability weight delta according to the results obtained in the step A3 and the step A4; the calculation formula of the step A5 is as follows:
wherein, delta is the weight,H max is the maximum value in matrix H;
a6, calculating a transmission rate estimation graph according to the atmospheric light value and the result of the step A4;
a7, carrying out weighted least square filtering processing on the transmission rate estimation graph according to the transmission rate credibility weight value; the specific process of the step A7 is as follows:
the following cost function is minimized:
wherein ,for matrix->Element of (a)>For matrix->Element N of (3) x A four-neighborhood representing a position x, y representing N x Coordinates of (E) above>I (y) represents N x Corresponding values are added;
the solution is as follows:
wherein ,for the corrected transmission rate estimation graph, delta represents the transmission rate reliability weight value obtained in the step A5, and ++>Representing the transmission rate estimation graph obtained in the step A6, D x 、D y Respectively converting the foggy images into gray level images, and then converting the foggy images into affinities of adjacent pixels in the x and y directions, wherein lambda is a smoothing coefficient, and T in the superscript represents transposition;
and A8, defogging the acquired fogged image by using the atmospheric light value and the transmission rate image processed in the step A7.
2. The improved dark channel defogging method using weighted least squares filtering according to claim 1, wherein the specific process of solving the cost function is:
a71, converting the foggy image into a gray scale image;
a72, calculating the affinity between adjacent pixels according to the gradient of the gray level map;
D x ＝-λ./[G(x+1，y)-G(x，y)]
D y ＝-λ./[G(x，y+1)-G(x，y)]
g (x, y) represents a gray value at (x, y);
a73, constructing a Laplacian homogeneous matrix by using the result in the step A72;
a74, adding the main diagonal elements of the Laplacian homogeneous matrix obtained in the step A73 with the elements in the weight delta respectively to obtain a parameter matrix M;
a75, element in weight delta and transmission rate estimation graphSequentially multiplying and expanding the elements in the sequence into a column vector N;
wherein N (x) represents an x-th element in N;
a76, multiplying the inverse point of the matrix M obtained in the step A74 by the column vector N obtained in the step A75;
V＝M -1 ·N
a77 step A76The result V is rearranged into a matrix with the same height and width as the image, and a corrected transmission rate diagram is obtained
3. The improved dark channel defogging method using weighted least squares filtering according to claim 2, wherein the formula of step A8 is:
where J (x) represents an image after defogging, and a represents an atmospheric light value.
4. The improved dark channel defogging method using weighted least squares filtering according to claim 3, wherein the formula of the matrix H in step A5 is replaced by:
wherein, omega (x) is a window taking x as a center r as a radius, a is the number of pixel points contained in the omega (x) window,and (3) obtaining a single-pixel dark channel diagram in the step A3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110249791.1A CN112927157B (en) | 2021-03-08 | 2021-03-08 | Improved dark channel defogging method adopting weighted least square filtering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110249791.1A CN112927157B (en) | 2021-03-08 | 2021-03-08 | Improved dark channel defogging method adopting weighted least square filtering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112927157A CN112927157A (en) | 2021-06-08 |
| CN112927157B true CN112927157B (en) | 2023-08-15 |
Family
ID=76171752
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110249791.1A Active CN112927157B (en) | 2021-03-08 | 2021-03-08 | Improved dark channel defogging method adopting weighted least square filtering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112927157B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108154482A (en) * | 2017-10-24 | 2018-06-12 | 天津大学 | Image Blind deblurring method based on dark channel prior and multi-direction Weighted T V |
| CN109064419A (en) * | 2018-07-12 | 2018-12-21 | 四川大学 | A kind of removing rain based on single image method based on WLS filtering and multiple dimensioned sparse expression |
| KR20190037096A (en) * | 2017-09-28 | 2019-04-05 | 가부시키가이샤 스크린 홀딩스 | Terahertz wave transmission measuring system and terahertz wave transmission measuring method |
| WO2019205707A1 (en) * | 2018-04-26 | 2019-10-31 | 长安大学 | Dark channel based image defogging method for linear self-adaptive improvement of global atmospheric light |
| CN110533619A (en) * | 2019-09-04 | 2019-12-03 | 河南工业大学 | A kind of image defogging improved method based on dark channel prior |
| CN110706298A (en) * | 2019-09-10 | 2020-01-17 | 天津大学 | Regularization weighted least square transmission-reflection dual-mode ultrasonic imaging reconstruction method |
| CN110717869A (en) * | 2019-09-11 | 2020-01-21 | 哈尔滨工程大学 | Underwater turbid image sharpening method |
| CN110827218A (en) * | 2019-10-31 | 2020-02-21 | 西北工业大学 | Airborne image defogging method based on image HSV transmissivity weighted correction |
| CN110910319A (en) * | 2019-10-30 | 2020-03-24 | 中国医学科学院北京协和医院 | Operation video real-time defogging enhancement method based on atmospheric scattering model |
| CN111489302A (en) * | 2020-03-27 | 2020-08-04 | 武汉理工大学 | Maritime image enhancement method in fog environment |
| WO2020206630A1 (en) * | 2019-04-10 | 2020-10-15 | 深圳市大疆创新科技有限公司 | Neural network for image restoration, and training and use method therefor |
| CN112419162A (en) * | 2019-08-20 | 2021-02-26 | 浙江宇视科技有限公司 | Image defogging method and device, electronic equipment and readable storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102104403B1 (en) * | 2013-05-28 | 2020-04-28 | 한화테크윈 주식회사 | Method and Apparatus for removing haze in a single image |
| US9361670B2 (en) * | 2014-09-04 | 2016-06-07 | National Taipei University Of Technology | Method and system for image haze removal based on hybrid dark channel prior |
| CN110232666B (en) * | 2019-06-17 | 2020-04-28 | 中国矿业大学(北京) | Underground pipeline image rapid defogging method based on dark channel prior |
-
2021
- 2021-03-08 CN CN202110249791.1A patent/CN112927157B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20190037096A (en) * | 2017-09-28 | 2019-04-05 | 가부시키가이샤 스크린 홀딩스 | Terahertz wave transmission measuring system and terahertz wave transmission measuring method |
| CN108154482A (en) * | 2017-10-24 | 2018-06-12 | 天津大学 | Image Blind deblurring method based on dark channel prior and multi-direction Weighted T V |
| WO2019205707A1 (en) * | 2018-04-26 | 2019-10-31 | 长安大学 | Dark channel based image defogging method for linear self-adaptive improvement of global atmospheric light |
| CN109064419A (en) * | 2018-07-12 | 2018-12-21 | 四川大学 | A kind of removing rain based on single image method based on WLS filtering and multiple dimensioned sparse expression |
| WO2020206630A1 (en) * | 2019-04-10 | 2020-10-15 | 深圳市大疆创新科技有限公司 | Neural network for image restoration, and training and use method therefor |
| CN112419162A (en) * | 2019-08-20 | 2021-02-26 | 浙江宇视科技有限公司 | Image defogging method and device, electronic equipment and readable storage medium |
| CN110533619A (en) * | 2019-09-04 | 2019-12-03 | 河南工业大学 | A kind of image defogging improved method based on dark channel prior |
| CN110706298A (en) * | 2019-09-10 | 2020-01-17 | 天津大学 | Regularization weighted least square transmission-reflection dual-mode ultrasonic imaging reconstruction method |
| CN110717869A (en) * | 2019-09-11 | 2020-01-21 | 哈尔滨工程大学 | Underwater turbid image sharpening method |
| CN110910319A (en) * | 2019-10-30 | 2020-03-24 | 中国医学科学院北京协和医院 | Operation video real-time defogging enhancement method based on atmospheric scattering model |
| CN110827218A (en) * | 2019-10-31 | 2020-02-21 | 西北工业大学 | Airborne image defogging method based on image HSV transmissivity weighted correction |
| CN111489302A (en) * | 2020-03-27 | 2020-08-04 | 武汉理工大学 | Maritime image enhancement method in fog environment |
Non-Patent Citations (1)
| Title |
|---|
| 李红云等."基于边界限制加权最小二乘法滤波的雾天图像增强方法".《中国激光》.2019,第46卷(第3期),第1-9页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112927157A (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107767354A (en) | A kind of image defogging algorithm based on dark primary priori | |
| CN102722868A (en) | Tone mapping method for high dynamic range image | |
| CN107705254B (en) | City environment assessment method based on street view | |
| US10609303B2 (en) | Method and apparatus for rapid improvement of smog/low-light-level image using mapping table | |
| CN109801233B (en) | Method for enhancing true color remote sensing image | |
| CN113298810A (en) | Trace detection method combining image enhancement and depth convolution neural network | |
| CN107862672B (en) | Image defogging method and device | |
| CN106934806A (en) | It is a kind of based on text structure without with reference to figure fuzzy region dividing method out of focus | |
| CN111754433B (en) | Defogging method for aerial image | |
| EP3407589B1 (en) | Image processing apparatus, image processing method, and storage medium | |
| JP2007531469A (en) | Hanging wire detection in color digital images | |
| CN105989583A (en) | Image defogging method | |
| CN112053298A (en) | Image defogging method | |
| CN114998122A (en) | Low-illumination image enhancement method | |
| CN111192213A (en) | Image defogging adaptive parameter calculation method, image defogging method and system | |
| US9083913B2 (en) | Image processing method, image processing device, and program | |
| CN112927157B (en) | Improved dark channel defogging method adopting weighted least square filtering | |
| CN111161159B (en) | Image defogging method and device based on combination of priori knowledge and deep learning | |
| CN107424134A (en) | Image processing method, device, computer-readable recording medium and computer equipment | |
| CN111241911A (en) | Self-adaptive lane line detection method | |
| CN108280809B (en) | Foggy image sky region estimation method based on atmospheric scattering physical model | |
| KR102439145B1 (en) | Method and system for Auto image dehazing using single scale image fusion | |
| CN107203979B (en) | Low-illumination image enhancement method | |
| CN109886901B (en) | Night image enhancement method based on multi-channel decomposition | |
| CN109801238B (en) | Weak illumination image enhancement method based on simplified atmospheric scattering model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |