CN113298729A - Rapid single image defogging method based on minimum value channel - Google Patents
Rapid single image defogging method based on minimum value channel Download PDFInfo
- Publication number
- CN113298729A CN113298729A CN202110563244.0A CN202110563244A CN113298729A CN 113298729 A CN113298729 A CN 113298729A CN 202110563244 A CN202110563244 A CN 202110563244A CN 113298729 A CN113298729 A CN 113298729A
- Authority
- CN
- China
- Prior art keywords
- image
- fog
- scene
- transmittance
- minimum value
- 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
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000002834 transmittance Methods 0.000 claims abstract description 27
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000001914 filtration Methods 0.000 claims abstract description 11
- 238000004364 calculation method Methods 0.000 claims abstract description 4
- 238000012887 quadratic function Methods 0.000 claims description 12
- 238000003384 imaging method Methods 0.000 claims description 11
- 238000009499 grossing Methods 0.000 claims description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 4
- 230000001105 regulatory effect Effects 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 10
- 230000008859 change Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000003595 mist Substances 0.000 description 2
- 241001669696 Butis Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 125000001475 halogen functional group Chemical group 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012880 independent component analysis Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000002311 subsequent effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
A rapid single image defogging method based on a minimum value channel relates to the technical field of digital image processing and solves the problems of inaccurate estimation of local transmittance and high calculation complexity in the existing defogging restoration method. The method utilizes the guide filtering to carry out smooth filtering processing on the transmittance image, namely, the transmittance can be accurately estimated, the algorithm processing effect is good, the operand is small, and the method can be applied to a real-time engineering system.
Description
Technical Field
The invention relates to the technical field of digital image processing, in particular to a rapid single image defogging method based on a minimum value channel.
Background
The visible light imaging system is affected by haze weather, so that image information acquired by the imaging sensor is seriously degraded, and the utility of outdoor imaging equipment is limited to a great extent. The application value of the image. How to improve the quality of haze degraded images and increase the visibility of the images is a leading topic of common concern in the fields of image processing and computer vision, and especially becomes a hot spot of current research aiming at the quality improvement of single images. Currently, solutions to such problems can be generalized to non-model and model based defogging methods. The non-model-based method mainly improves the visual effect of the image in foggy days, such as histogram equalization, gamma correction, and Retinex algorithm based on color constancy, and although such algorithms are simple and fast, the loss of other information is usually caused while the detailed information of the image is enhanced. The defogging method based on the physical model is established based on an imaging degradation model, and the non-fitness of the atmospheric transmission and the environmental illumination estimation solution in the atmospheric scattering model is converted into the fitness through a plurality of prior information, such as dark primary color prior, geometric prior and the like.
He and the like put forward a dark primary color principle through statistics of outdoor scene images, and realize recovery of the scene through solving a scene dark channel, and the algorithm obtains a good defogging effect, but has high complexity; fattal assumes that the albedo of a local region in an image is a constant, and independent component analysis is adopted to solve the scene albedo, but the influence of image statistical characteristics on algorithm processing results is large; tan utilizes a Markov random field to restrain the image contrast and other characteristics of the outdoor scene according to the prior assumption of the image contrast and the like of the outdoor scene, solves the atmospheric light attenuation term in the model, can better recover the scene information in the image, but the color of the recovered scene is easy to have the oversaturation phenomenon. It can be seen that when some prior assumptions are not satisfied, a significant deviation occurs in the estimation of the local atmospheric transmittance, which causes color distortion, unclear restoration of image details and halo phenomenon in the defogged image. In addition, the high complexity of the algorithm also limits the application of the algorithm in practical engineering.
Disclosure of Invention
The invention provides a rapid single image defogging method based on a minimum channel, aiming at solving the problems of inaccurate estimation of local transmittance and high calculation complexity in the existing defogging restoration method.
A rapid single image defogging method based on a minimum value channel is realized by the following steps:
step one, carrying out transmittance estimation on an imaging model based on atmospheric scattering to obtain a transmittance estimation valueRepresented by the formula:
in the formula,for the minimum channel image of a foggy scene,the atmospheric light value is the atmospheric light value of the minimum channel, c is one color channel in the r, g and b channels, and omega and sigma are adjusting constants;the minimum channel image of the fog-free scene is obtained;
step two, adopting the minimum value channel image of the foggy sceneMinimum value channel image for fog-free sceneThe estimation is carried out, namely:andthe relationship is expressed as a quadratic function combination by the following formula:
in the formula, the vertex of the quadratic function is (I)ave,n),IaveMean of foggy images, m1And m2Is a quadratic coefficient;
step three, selectingThe first 0.1% of the maximum value of (c) as the region omega of the atmospheric light value calculationJ,The total pixel number is N, and a corresponding area foggy image I is calculatedc(x) Has a mean value of the gray levels of three channels of AcThen, the gray level average A of the three channels is obtainedcIs taken as the atmospheric light valueIs expressed by the following formula:
step four, obtaining the minimum value channel image of the fog-free scene according to the quadratic function combination form in the step twoWill be provided withSubstituting the transmittance estimated value into the first stepObtaining an estimated value of transmittanceSmoothing by adopting smoothing filtering to obtain the transmittance t (x);
and step five, restoring the fog-free scene according to an imaging model of atmospheric scattering, wherein the fog-free scene is expressed by the following formula:
in the formula, t0For the adjustment amount, i (x) is a fog scene image, and j (x) is a fog-free scene image.
The invention has the beneficial effects that: the method of the invention establishes an estimation model of the transmittance through the fog image and the fog-free image to the minimum channel, and obtains the minimum channel of the fog-free image through a quadratic function transformation model, thereby obtaining a transmittance map.
The method utilizes the guide filtering to carry out smooth filtering processing on the transmittance image, namely, the transmittance can be accurately estimated, the algorithm processing effect is good, the operand is small, and the method can be applied to a real-time engineering system.
Drawings
FIG. 1 is a graph of z versus y when n takes on different values;
FIG. 2 is IaveWhen different values are taken, a relation graph of z and y is obtained;
FIG. 3 is a diagram showing the effect of the defogging process; wherein, (a) is the minimum value channel image of the foggy sceneThe (b) is a minimum value channel image of the fog-free sceneThe effect graph of (c) is a transmittance estimated valueThe subsequent effect graph, (d) is an effect graph of transmittance t (x) after guiding filtering smoothing;
FIG. 4 is a diagram showing the effect of the defogging treatment; wherein, (a) is the original image, (b) is the effect picture of the dark channel prior result, and (c) is the effect picture processed by the method of the invention.
Detailed Description
The embodiment is described with reference to fig. 1 to 4, and a method for defogging a single image based on a minimum value channel includes:
1. atmospheric scattering model:
at present, the widely used fog imaging model is derived from narasimahan and Nayar based on McCartney attenuation model and atmospheric light model, and the atmospheric scattering based imaging model is shown as follows:
I(x)=J(x)t(x)+A(1-t(x)) (1)
wherein, i (x) is a foggy scene image, j (x) is a fogless scene image, a is global atmospheric light, and t (x) is atmospheric transmittance.
2. Establishing a transmittance estimation model:
performing minimum value filtering on the local region of equation (1) to obtain a rough estimation of transmittance:
wherein,for the minimum channel image of a foggy scene,the atmospheric light values for the minimum channel, ω and σ, are tuning constants.Is a minimum value channel image of a fog-free scene, butIs an unknown quantity, and we will now describe the use of a known quantityFor unknown quantityA method of performing the estimation.
3. Establishing a nonlinear model:
will be provided withAndcarrying out normalization processing, and under the fog condition, carrying out the minimum value channel of the fog-free imageFollowing a foggy imageIs enlarged by increasing, and is provided withTherefore, in the present embodiment, a non-linear increasing function relationship is assumed to exist between the two.
In general, a hazy image increases with scene depth,becoming progressively larger. Through a large number of experiments, it is found thatSmaller, indicating less affected by the fog,the speed increasing gradually becomes smaller, namely the slope of the curve becomes smaller; mean value of fog image is IaveWhen is coming into contact withWhen the position of the mobile phone is close to the base station,the speed increasing is minimum, namely the slope of the curve is minimum; followed byIs facingThe increase in the amount of the mist indicates that the effect of the mist is increased,the increasing speed is gradually larger, namely the slope of the curve is increased. As can be seen,andis a piecewise functional relationship, given in this embodimentAndthe relation is shown in formula (3).
The formula (3) is a quadratic function combination form, and the vertex of the piecewise quadratic function is (I)aveN). Because of the fact thatBy using the boundary condition, whenWhen the temperature of the water is higher than the set temperature,m is obtained from the formula (3)1=-n/(Iave)2. When in useWhen the temperature of the water is higher than the set temperature,m is obtained from the formula (3)2=(1-n)/(1-Iave)2。
From the above analysis, m1>0,m2<0。m1And m2The coefficient is a quadratic term coefficient, the slope change of the control curve is larger, the rising speed of the curve is higher, and the slope change is larger.
when in useWhen the slope takes the maximum value, kmax=2n/IaveFrom the above analysis, it can be seen that the slope should be less than 1, and that n ≦ Iave/2. When I isaveWhen n is 0.5, n is 0.1, 0.2, 0.25, 0.3, and the model established by the formula (3) shows that n is 0.1, 0.2, 0.25, 0.3The corresponding z to y relationship is then as shown in figure 1. Average value of general foggy image IaveThe larger the fog density of the imaging scene; conversely, the fog density of the imaged scene is smaller. When the value of n is smaller, the area enclosed by the curve and the straight line y which is z is larger, and the method is suitable for processing the image with larger fog concentration; conversely, when the value of n is larger, the area enclosed by the curve and the straight line y ═ z is smaller, and the image with a smaller fog density is suitable for processing. Then n and IaveThe relationship of (1): n ═ Iave/2-0.05. When I isaveValues are respectively Iave=0.3、Iave=0.4、Iave=0.5、Iave0.6 is givenThen, corresponding z and yThe relationship is shown in fig. 2.
4. Obtaining the transmittance;
selectingThe area in which the first 0.1% of the pixels of the maximum value of (A) are calculated as the atmospheric light value isThe total number of pixels is N, and the corresponding fog image I is calculatedc(x) Has a mean value of the gray levels of three channels of AcThen, three channels A are obtainedcIs taken as the atmospheric light valueThe expression of (c) is as follows.
5. Restoring the image;
according to the model established by the formula (3), the minimum channel of the fog-free scene of the image can be obtainedWill be provided withAnd the equation (5) is substituted into the transmittance estimation model equation (2) to obtain the transmittance estimation valueThen, smoothing (such as guiding filtering, Gaussian filtering, bilateral filtering, etc.) is performed to obtain t (x). And then, restoring the fog-free scene according to an atmospheric scattering model, wherein the formula is as follows:
in the formula, t0In the present embodiment, the adjustment amount is set to 0.01. The image restored by the above method is generally darker and is enhanced by an automatic color gradation method. The processing results are shown in fig. 4.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (3)
1. The rapid single image defogging method based on the minimum value channel is characterized by comprising the following steps: the method is realized by the following steps:
step one, carrying out transmittance estimation on an imaging model based on atmospheric scattering to obtain a transmittance estimation valueRepresented by the formula:
in the formula,for the minimum channel image of a foggy scene,is the maximum of the minimum value channelC is one color channel in r, g and b channels, and omega and sigma are regulating constants;the minimum channel image of the fog-free scene is obtained;
step two, adopting the minimum value channel image of the foggy sceneMinimum value channel image for fog-free sceneThe estimation is carried out, namely:andthe relationship is expressed as a quadratic function combination by the following formula:
in the formula, the vertex of the quadratic function is (I)ave,n),eIvaMean of foggy images, m1And m2Is a quadratic coefficient;
step three, selectingThe first 0.1% of the maximum value of (c) as the region omega of the atmospheric light value calculationJ,The total pixel number is N, and a corresponding area foggy image I is calculatedc(x) Has a mean value of the gray levels of three channels of AcThen, the gray level average A of the three channels is obtainedcIs taken as the atmospheric light valueIs expressed by the following formula:
step four, obtaining the minimum value channel image of the fog-free scene according to the quadratic function combination form in the step twoWill be provided withSubstituting the transmittance estimated value into the first stepObtaining an estimated value of transmittanceSmoothing by adopting smoothing filtering to obtain the transmittance t (x);
and step five, restoring the fog-free scene according to an imaging model of atmospheric scattering, wherein the fog-free scene is expressed by the following formula:
in the formula, t0For the adjustment amount, i (x) is a fog scene image, and j (x) is a fog-free scene image.
2. The method for defogging a single image based on a minimum value channel according to claim 1, wherein: in the second step:
when in useWhen the temperature of the water is higher than the set temperature,m can be obtained according to a quadratic function formula1=-n/(Iave)2;
When in useWhen the temperature of the water is higher than the set temperature,can obtain m2=(1-n)/(1-Iave)2;
Finally, the following can be obtained: m is1>0,m2<0;
3. The method for defogging a single image based on a minimum value channel according to claim 1, wherein: and step five, performing enhancement processing on the restored image by adopting an automatic color gradation method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110563244.0A CN113298729B (en) | 2021-05-24 | 2021-05-24 | Rapid single image defogging method based on minimum value channel |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110563244.0A CN113298729B (en) | 2021-05-24 | 2021-05-24 | Rapid single image defogging method based on minimum value channel |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113298729A true CN113298729A (en) | 2021-08-24 |
CN113298729B CN113298729B (en) | 2022-04-26 |
Family
ID=77324053
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110563244.0A Active CN113298729B (en) | 2021-05-24 | 2021-05-24 | Rapid single image defogging method based on minimum value channel |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113298729B (en) |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942758A (en) * | 2014-04-04 | 2014-07-23 | 中国人民解放军国防科学技术大学 | Dark channel prior image dehazing method based on multiscale fusion |
CN105957040A (en) * | 2016-05-19 | 2016-09-21 | 湖南源信光电科技有限公司 | Rapid defog algorithm based on image fusion |
CN106157270A (en) * | 2016-08-29 | 2016-11-23 | 潍坊学院 | A kind of single image rapid defogging method and system |
CN106846263A (en) * | 2016-12-28 | 2017-06-13 | 中国科学院长春光学精密机械与物理研究所 | The image defogging method being immunized based on fusion passage and to sky |
KR101784350B1 (en) * | 2016-06-02 | 2017-10-11 | 한국항공대학교산학협력단 | Dehazing method and device based on improved median dark channel prior |
CN109767407A (en) * | 2019-02-27 | 2019-05-17 | 长安大学 | A kind of quadratic estimate method of atmospheric transmissivity image during defogging |
CN111598886A (en) * | 2020-05-25 | 2020-08-28 | 中国科学院长春光学精密机械与物理研究所 | Pixel-level transmittance estimation method based on single image |
CN111598791A (en) * | 2020-04-13 | 2020-08-28 | 西安理工大学 | Image defogging method based on improved dynamic atmospheric scattering coefficient function |
CN111598812A (en) * | 2020-05-25 | 2020-08-28 | 中国科学院长春光学精密机械与物理研究所 | Image defogging method based on RGB and HSV double-color space |
US20200394767A1 (en) * | 2019-06-17 | 2020-12-17 | China University Of Mining & Technology, Beijing | Method for rapidly dehazing underground pipeline image based on dark channel prior |
CN113298730A (en) * | 2021-05-24 | 2021-08-24 | 中国科学院长春光学精密机械与物理研究所 | Defogging restoration method based on image decomposition |
-
2021
- 2021-05-24 CN CN202110563244.0A patent/CN113298729B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942758A (en) * | 2014-04-04 | 2014-07-23 | 中国人民解放军国防科学技术大学 | Dark channel prior image dehazing method based on multiscale fusion |
CN105957040A (en) * | 2016-05-19 | 2016-09-21 | 湖南源信光电科技有限公司 | Rapid defog algorithm based on image fusion |
KR101784350B1 (en) * | 2016-06-02 | 2017-10-11 | 한국항공대학교산학협력단 | Dehazing method and device based on improved median dark channel prior |
CN106157270A (en) * | 2016-08-29 | 2016-11-23 | 潍坊学院 | A kind of single image rapid defogging method and system |
CN106846263A (en) * | 2016-12-28 | 2017-06-13 | 中国科学院长春光学精密机械与物理研究所 | The image defogging method being immunized based on fusion passage and to sky |
CN109767407A (en) * | 2019-02-27 | 2019-05-17 | 长安大学 | A kind of quadratic estimate method of atmospheric transmissivity image during defogging |
US20200394767A1 (en) * | 2019-06-17 | 2020-12-17 | China University Of Mining & Technology, Beijing | Method for rapidly dehazing underground pipeline image based on dark channel prior |
CN111598791A (en) * | 2020-04-13 | 2020-08-28 | 西安理工大学 | Image defogging method based on improved dynamic atmospheric scattering coefficient function |
CN111598886A (en) * | 2020-05-25 | 2020-08-28 | 中国科学院长春光学精密机械与物理研究所 | Pixel-level transmittance estimation method based on single image |
CN111598812A (en) * | 2020-05-25 | 2020-08-28 | 中国科学院长春光学精密机械与物理研究所 | Image defogging method based on RGB and HSV double-color space |
CN113298730A (en) * | 2021-05-24 | 2021-08-24 | 中国科学院长春光学精密机械与物理研究所 | Defogging restoration method based on image decomposition |
Non-Patent Citations (8)
Title |
---|
SURESH CHANDRA RAIKWAR等: "Lower Bound on Transmission Using Non-Linear Bounding Function in Single Image Dehazing", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
李一菲: "一种自适应线性透射率估计去雾算法", 《应用光学》 * |
李一菲: "基于自适应透射率和Retinex理论的单幅图像去雾算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
李一菲等: "一种自适应的多级透射率估计去雾算法", 《光子学报》 * |
杨燕等: "基于区间估计与透射率自适应约束的去雾算法", 《北京航空航天大学学报》 * |
杨燕等: "基于相对透射率估计的快速图像去雾算法", 《计算机应用》 * |
毕国玲等: "基于暗通道图像质心偏移量的去雾算法", 《光学学报》 * |
毕国玲等: "基于照射反射模型和有界运算的多谱段图像增强", 《物理学报》 * |
Also Published As
Publication number | Publication date |
---|---|
CN113298729B (en) | 2022-04-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103218778B (en) | The disposal route of a kind of image and video and device | |
Gao et al. | Sand-dust image restoration based on reversing the blue channel prior | |
CN112419162B (en) | Image defogging method, device, electronic equipment and readable storage medium | |
Wang et al. | A fast single-image dehazing method based on a physical model and gray projection | |
CN109816605B (en) | MSRCR image defogging method based on multi-channel convolution | |
CN110782407B (en) | Single image defogging method based on sky region probability segmentation | |
CN108133462B (en) | Single image restoration method based on gradient field region segmentation | |
CN104182943B (en) | A kind of single image defogging method capable merging human-eye visual characteristic | |
CN111861896A (en) | UUV-oriented underwater image color compensation and recovery method | |
CN111598886B (en) | Pixel-level transmittance estimation method based on single image | |
Kumari et al. | Single image fog removal using gamma transformation and median filtering | |
CN104331867B (en) | The method, device and mobile terminal of image defogging | |
CN111192205A (en) | Image defogging method and system and computer readable storage medium | |
CN105023246B (en) | A kind of image enchancing method based on contrast and structural similarity | |
CN109949239B (en) | Self-adaptive sharpening method suitable for multi-concentration multi-scene haze image | |
CN114119383B (en) | Underwater image restoration method based on multi-feature fusion | |
CN108765337B (en) | Single color image defogging processing method based on dark channel prior and non-local MTV model | |
CN107977941B (en) | Image defogging method for color fidelity and contrast enhancement of bright area | |
CN114693548A (en) | Dark channel defogging method based on bright area detection | |
CN112419163A (en) | Single image weak supervision defogging method based on priori knowledge and deep learning | |
CN110852971B (en) | Video defogging method based on dark channel prior and Retinex and storage medium | |
Zhang et al. | A fast video image defogging algorithm based on dark channel prior | |
CN107451962B (en) | Image defogging method and device | |
CN112825189B (en) | Image defogging method and related equipment | |
CN113298729B (en) | Rapid single image defogging method based on minimum value channel |
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 |