CN115861104A - Remote sensing image defogging method based on transmissivity refinement - Google Patents
Remote sensing image defogging method based on transmissivity refinement Download PDFInfo
- Publication number
- CN115861104A CN115861104A CN202211544780.7A CN202211544780A CN115861104A CN 115861104 A CN115861104 A CN 115861104A CN 202211544780 A CN202211544780 A CN 202211544780A CN 115861104 A CN115861104 A CN 115861104A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- image
- sensing image
- foggy
- transmissivity
- 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 39
- 230000000877 morphologic effect Effects 0.000 claims description 7
- 230000010339 dilation Effects 0.000 claims description 3
- 230000003628 erosive effect Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract 1
- 238000004088 simulation Methods 0.000 description 11
- 238000002834 transmittance Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 125000001475 halogen functional group Chemical group 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 230000004927 fusion Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000007797 corrosion Effects 0.000 description 3
- 238000005260 corrosion Methods 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Abstract
The invention provides a remote sensing image defogging method based on transmissivity refinement, which is used for solving the problems of inaccurate estimation of transmissivity in a remote sensing image and low defogging quality of the remote sensing image; and the problem of low contrast of the remote sensing image after defogging. The method comprises the following implementation steps: calculating an atmospheric intensity value of the foggy remote sensing image; obtaining a Gaussian weighted dark channel image of the foggy remote sensing image by using a Gaussian weighted algorithm; calculating the thinning transmissivity of the foggy remote sensing image; and defogging the foggy remote sensing image according to the atmospheric intensity value and the refined transmissivity to obtain a defogged image. The invention overcomes the problem of inaccurate transmissivity calculation in the prior art, can carry out high-quality defogging on the foggy images including the remote sensing images, can improve the contrast of the images after defogging and increase the detail information of the images.
Description
Technical Field
The invention belongs to the technical field of image processing, and further relates to a remote sensing image defogging method based on transmissivity refinement in the technical field of remote sensing image processing. According to the invention, the degree of distortion of the remote sensing image and the visibility of an object in the remote sensing image can be reduced by eliminating fog and cloud in the remote sensing image, and the application scene of subsequent segmentation and target detection and identification of the remote sensing image is facilitated.
Background
When fog is present in the air, the acquired remote sensing images often have low contrast and poor visibility, which not only reduces the visual effect, but also hinders subsequent processing in computer vision systems. In response to this problem, a number of image defogging methods and techniques have been proposed for restoring remote sensing images. Common methods include a series of algorithms derived by taking a Retinex algorithm as a representative, histogram equalization, wavelet transformation and the like, but these methods are prone to cause loss of detail information, noise addition, and distortion problems such as supersaturation and serious loss of image detail information when an image is defogged. With the development of deep learning, a plurality of defogging methods based on various neural networks are proposed. However, image defogging is a step in image preprocessing and is not always the final desired result, and it is wasteful to spend a lot of time training a defogging neural network model.
A remote sensing image defogging method based on a single atmospheric scattering model is disclosed in patent document 'a foggy remote sensing image restoration method' (application number: 2019107303597, application publication number: CN 110490821A) applied by Changchun optical precision machinery and physical research institute of China academy of sciences. Firstly, establishing a remote sensing image degradation model based on atmospheric micro point light source active illumination on the basis of a single atmospheric scattering model, and acquiring a halo component of an original foggy remote sensing image through the remote sensing image degradation model; secondly, removing the atmosphere multi-scattering effect and non-uniformity of the original fog remote sensing image according to the halo component, and dividing the original fog remote sensing image into a halo image and a haze image after the halo is removed; and thirdly, estimating the transmittance and atmospheric light of the haze image with the halo removed by adopting a single atmospheric scattering model, and performing defogging restoration processing on the haze image with the halo removed by adopting a dark channel prior image defogging algorithm to obtain a defogging restoration image. The method simply realizes remote sensing image defogging. However, the method still has the defects that the acquired transmittance value is inaccurate and the defogging quality of the remote sensing image is poor due to the fact that the transmittance is not thinned.
Disclosure of Invention
The invention aims to provide a remote sensing image defogging method based on transmissivity refinement aiming at the defects of the prior art, and the remote sensing image defogging method is used for solving the problems that the transmissivity refinement is not carried out and the defogging quality of a remote sensing image is low in the prior art.
The technical idea for realizing the purpose of the invention is as follows: according to the method, the Gaussian weighting algorithm is used to obtain the Gaussian weighted dark channel image of the remote sensing image, the image transmissivity is refined, and the problem of inaccurate transmissivity value is solved. And a Gaussian weighting algorithm is used to obtain a finer transmission map, so that a finer transmittance can be obtained, and the problem of low defogging quality of the remote sensing image is solved.
In order to achieve the purpose, the technical scheme adopted by the invention comprises the following steps:
step 1, calculating an atmospheric intensity value A of a foggy remote sensing image I:
step 2.1, calculating a dark channel image I of the foggy remote sensing image I according to the following formula 1 ：
wherein ,
representing computationsThe minimum value of the R, G and B channels respectively represents three channels corresponding to red, green and blue pixels in the fog remote sensing image;
step 2.2, with I 2 ＝dilate(I 1 -erode(I 1 ) ) formula to obtain a dark channel image I 1 Edge image I of 2 Wherein, dilate represents morphological dilation operation and erode represents morphological erosion operation;
step 2.3, with I 3 ＝0.5I 2 +0.5I 1 Formula, calculating the compensated dark channel image I of the foggy remote sensing image I 3 ；
Step 2.4, calculating a Gaussian weighted dark channel image I' of the fog remote sensing image I according to the following formula:
wherein ,e{· Denotes exponential operation with a natural constant e as the base;
step 3, calculating the refined transmittance T of the foggy remote sensing image I by using a formula T = 255-I';
step 4, defogging the foggy remote sensing image I according to the atmospheric intensity value A and the refined transmissivity T to obtain a defogged image R:
wherein, F represents a center surround function,
a (1-T) represents an atmospheric light attenuating portion, based on the measured value of the intensity of the light in the atmosphere>
Representing a convolution operation.
Compared with the prior art, the invention has the following advantages:
the Gaussian weighted dark channel image of the foggy remote sensing image is obtained through an image Gaussian weighted fusion algorithm, the refined transmittance of the foggy remote sensing image is calculated, the problem that the transmittance in the prior art is not accurately calculated is solved, and the defogging quality of the remote sensing image is improved.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a simulation of the present invention, wherein FIG. 2 (a) is an original image containing fog, FIG. 2 (b) is a defogged image of the original image 2 (a) by a prior art method, and FIG. 2 (c) is a defogged image of the original image 2 (a) by a method of the present invention;
Detailed Description
The invention is described in further detail below with reference to the figures and specific examples.
Referring to fig. 1, the present invention includes the steps of:
step 1, obtaining an atmospheric intensity value A of a foggy image I.
Step 1.1, the brightness of pixels in an image is between 0 and 255, the brightness of pixels close to 255 is high, and the brightness close to 0 is low; all pixels in the foggy remote sensing image I with the size of L multiplied by W are sorted from large to small according to the brightness to form a brightness pixel set N, wherein L is more than or equal to 512, and W is more than or equal to 512; selecting the first 0.1% of pixels from the luminance pixel set N to form the luminance pixel set N 1 ；
Step 1.2, in the brightness pixel set N 1 In (3), the brightness value of the first pixel is counted as B 1 The brightness value of the second pixel is B 2 The brightness value of the third pixel is counted as B 3 By analogy, the luminance value of the ith pixel is counted as
wherein ,/>
Represents a round-down operation; set the luminance pixels N 1 The brightness values of all the pixels are summed to obtain a brightness value sum B sum ，/>
Wherein Σ denotes a summation operation;
step 1.3, calculating a brightness pixel set N 1 The average value of the brightness values of all the pixels in the image is used for obtaining the atmospheric intensity value A of the foggy remote sensing image I,
and 2, obtaining a Gaussian weighted dark channel image I' of the foggy remote sensing image I by using a Gaussian weighted algorithm.
Step 2.1, a dark channel priori principle is established on the basis of observation of a large number of outdoor fog-free images, and the dark channel priori assumption is that the value of a color channel tends to be 0 in some pixels in most non-sky local areas; obtaining a dark channel image I of the foggy remote sensing image I according to the dark channel prior principle 1 Can be represented as
wherein ,/>
The minimum value of R, G and B channels is calculated, and R, G and B respectively represent three channels corresponding to red, green and blue pixels in the fog remote sensing image;
step 2.2, with I 2 ＝dilate(I 1 -erode(I 1 ) ) formula to obtain a dark channel image I 1 Edge image I of 2 Wherein, dilate represents morphological dilation operation and erode represents morphological erosion operation;
the corrosion operation of the image can eliminate the boundary point of the image, so that the image shrinks inwards along the boundary, and the part smaller than the specified structural element can be removed, so that the noise in the image can be effectively eliminated; the expansion operation of the image is opposite to the corrosion operation, the expansion operation can expand the boundary of the image, and the boundary information of the image is accurately extracted; the method comprises the following steps that firstly, the corrosion and then the expansion form a morphological opening operation, so that the noise of the image can be effectively eliminated, and meanwhile, the edge information of the image can be accurately positioned;
step 2.3, with I 3 ＝0.5I 2 +0.5I 1 Formula, calculating the compensated dark channel image I of the foggy remote sensing image I 3 ；
The image weighted fusion operation can enlarge the time space information contained in the image, reduce the uncertainty and increase the reliability; edge image I 2 And dark channel image I 1 The dark channel pixels at the edge of the image can be effectively compensated by carrying out weighted fusion;
step 2.4, calculating a Gaussian weighted dark channel image I' of the fog remote sensing image I according to the following formula:
wherein ,e{· Denotes exponential operation with a natural constant e as the base;
the gaussian weighting function is a process of performing weighted average on the whole image, and the value of each pixel point is obtained by performing weighted average on the value of each pixel point and other pixel values in the neighborhood. The Gaussian weighting function can be used for effectively refining the pixel boundary of the dark channel image, and an image with finer pixel edge is obtained.
Step 3, calculating the refined transmittance T of the foggy remote sensing image I by using a formula T = 255-I';
and 4, obtaining a defogged image R of the fogging remote sensing image I.
Step 4.1, an atmospheric scattering model table is I = Jt + A (1-t), wherein I represents a foggy remote sensing image, J represents a fogless clear remote sensing image, t represents a transmissivity, and A represents an atmospheric intensity value; the Retinex theoretical model is denoted S = RL, where S denotes a fog-free sharp remote sensing image, R denotes a reflection component, L denotes an illumination component, where,
representing a convolution operation, F representing a center surround function, based on a convolution function>
Step 4.2, fusing the atmospheric scattering model and the Retinex theoretical model, and if J = S, then fusing the defogging model as
Step 4.2, according to the fusion defogging model, combining the atmospheric intensity value A and the refined transmissivity T, defogging the foggy remote sensing image I to obtain a defogged image R:
wherein, F represents a center surround function,
a (1-T) represents an atmospheric light attenuating portion, based on the measured value of the intensity of the light in the atmosphere>
Representing a convolution operation.
The effect of the present invention is further explained by combining the simulation experiment as follows:
1. simulation experiment conditions are as follows:
simulation experiment of the invention the hardware platform of the experiment is: the processor is Intel i5-10400F, the main frequency is 2.9GHz, and the 1lg running memory.
The software platform of the simulation experiment platform of the invention is as follows: windows 11 operating system and Visual Studio2017 software.
The image used in the simulation experiment of the present invention is from the paper "A remote sensing image dataset for closed removal" (Computer Vision and Pattern Recognition, vol. Abs/1901.00600).
2. Simulation content and result analysis thereof:
the simulation experiment of the invention adopts the invention and the prior art (a new defogging method for the remote sensing image of the unmanned aerial vehicle) to respectively perform defogging treatment on a foggy remote sensing image, and the result is shown in figure 2.
In the simulation experiment, one prior art adopted refers to:
tao Gao et al, in its publication, "A novel UAV Sensing Image Defogging Method" (IEEE Journal of Selected Topics in Applied Earth objects being used for Defogging and Remote Sensing images, 2020, pp.2610-2625).
The simulation effect of the present invention is further described below with reference to fig. 2.
Fig. 2 (a) is an original image containing fog, fig. 2 (b) is a defogged image of the original image like fig. 2 (a) by using a method of the prior art, and fig. 2 (c) is a defogged image of the original image 2 (a) by using a method of the present invention.
As can be seen from fig. 2 (b), the residual fog still exists in the image after the existing method defogging, and the edge of the river in fig. 2 (b) is blurred, and the ground object is not clearly displayed, so the existing method defogging quality is not high, and the detail recovery is poor. Compared with the prior method. In the step of fig. 2 (c), the residual fog is completely eliminated at the abrupt change of the depth of field, the color cast phenomenon is inhibited, the detail information of the ground object can be retained, and the defogging effect is more prominent. Therefore, the defogging effect of the image in the figure 2 (c) is good, the whole color is natural, the haze in the image can be well removed, and the image contrast and the saturation are improved.
Comparing the invention with the existing published classical methods, the comparison methods are respectively as follows: DCP, CAP and MSBDN; the adopted evaluation index is Structural Similarity (SSIM), the range of SSIM is 0-1, and the higher the value is, the higher the defogged image quality is. Referring to fig. 2 (a), SSIM of DCP method is 0.405, SSIM of cap method is 0.774, SSIM of msbdn method is 0.647, SSIM of the inventive method is 0.873. Therefore, the remote sensing image subjected to defogging by the method has natural integral color, higher image contrast and higher defogging quality.
The simulation experiment results show that the method has good defogging capability on the remote sensing image containing fog; the invention can accurately recover the image and has high defogging quality.
Claims (2)
1. A remote sensing image defogging method based on transmissivity thinning is characterized in that a Gaussian weighting algorithm is used for obtaining a Gaussian weighted dark channel image of a foggy remote sensing image, and thinning transmissivity is calculated; the defogging method comprises the following steps:
step 1, calculating an atmospheric intensity value A of a foggy remote sensing image I:
step 2, obtaining a Gaussian weighted dark channel image I' of the foggy remote sensing image I by using a Gaussian weighted algorithm:
step 2.1, calculating a dark channel image I of the foggy remote sensing image I according to the following formula 1 ：
wherein ,
the minimum value of R, G and B channels is calculated, and R, G and B respectively represent three channels corresponding to red, green and blue pixels in the fog remote sensing image;
step 2.2, with I 2 ＝dilate(I 1 -erode(I 1 ) ) formula to obtain a dark channel image I 1 Edge image I of 2 Wherein, dilate represents morphological dilation operation and erode represents morphological erosion operation;
step 2.3, with I 3 ＝0.5I 2 +0.5I 1 Formula, calculating the compensated dark channel image I of the foggy remote sensing image I 3 ；
Step 2.4, calculating a Gaussian weighted dark channel image I' of the fog remote sensing image I according to the following formula:
wherein ,e{· Denotes an exponential operation with a natural constant e as the baseMaking;
step 3, calculating the thinning transmissivity T of the foggy remote sensing image I by using a formula T = 255-I';
step 4, defogging the foggy remote sensing image I according to the atmospheric intensity value A and the refined transmissivity T to obtain a defogged image R:
wherein, F represents a center surround function,
a (1-T) represents an atmospheric light attenuation section, based on a light intensity value of the atmosphere>
Representing a convolution operation.
2. The remote sensing image defogging method based on transmissivity refinement of claim 1, wherein said atmospheric intensity value A of step 1 is obtained by the following steps:
firstly, the brightness of pixels in an image is between 0 and 255, the brightness of pixels close to 255 is high, and the brightness close to 0 is low; all pixels in the foggy remote sensing image I with the size of L multiplied by W are sorted from large to small according to the brightness to form a brightness pixel set N, wherein L is more than or equal to 512, and W is more than or equal to 512; selecting the first 0.1% of pixels from the luminance pixel set N to form the luminance pixel set N 1 ；
Second, in the set of luminance pixels N 1 In (3), the brightness value of the first pixel is counted as B 1 The brightness value of the second pixel is B 2 The brightness value of the third pixel is B 3 By analogy, the luminance value of the ith pixel is counted as B i ，
wherein ,/>
Represents a round-down operation; set the luminance pixels N 1 The brightness values of all the pixels are summed to obtain a brightness value sum B sum ，/>
Wherein Σ denotes a summation operation;
thirdly, calculating a brightness pixel set N 1 The average value of the brightness values of all the pixels in the image is used for obtaining the atmospheric intensity value A of the foggy remote sensing image I,
/>
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211544780.7A CN115861104B (en) | 2022-11-30 | 2022-11-30 | Remote sensing image defogging method based on transmissivity refinement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211544780.7A CN115861104B (en) | 2022-11-30 | 2022-11-30 | Remote sensing image defogging method based on transmissivity refinement |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115861104A true CN115861104A (en) | 2023-03-28 |
| CN115861104B CN115861104B (en) | 2023-10-17 |
Family
ID=85669687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211544780.7A Active CN115861104B (en) | 2022-11-30 | 2022-11-30 | Remote sensing image defogging method based on transmissivity refinement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115861104B (en) |
Citations (9)
| 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 |
| KR101426298B1 (en) * | 2014-01-24 | 2014-08-13 | (주)인펙비전 | apparatus and method for compensating image for enhancing fog removing efficiency |
| KR101448164B1 (en) * | 2013-04-22 | 2014-10-14 | 금오공과대학교 산학협력단 | Method for Image Haze Removal Using Parameter Optimization |
| KR20150002187A (en) * | 2013-06-28 | 2015-01-07 | 주식회사 시큐인포 | System and method for image defogging based on gaussian filtering |
| US20160071244A1 (en) * | 2014-09-04 | 2016-03-10 | National Taipei University Of Technology | Method and system for image haze removal based on hybrid dark channel prior |
| CN112465708A (en) * | 2020-10-23 | 2021-03-09 | 南京理工大学 | Improved image defogging method based on dark channel |
| CN113487509A (en) * | 2021-07-14 | 2021-10-08 | 杭州电子科技大学 | Remote sensing image fog removing method based on pixel clustering and transmissivity fusion |
| CN114119411A (en) * | 2021-11-24 | 2022-03-01 | 湖南中科助英智能科技研究院有限公司 | Fog noise video image recovery method, device, equipment and medium |
| CN114219732A (en) * | 2021-12-15 | 2022-03-22 | 大连海事大学 | Image defogging method and system based on sky region segmentation and transmissivity refinement |
-
2022
- 2022-11-30 CN CN202211544780.7A patent/CN115861104B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101448164B1 (en) * | 2013-04-22 | 2014-10-14 | 금오공과대학교 산학협력단 | Method for Image Haze Removal Using Parameter Optimization |
| KR20150002187A (en) * | 2013-06-28 | 2015-01-07 | 주식회사 시큐인포 | System and method for image defogging based on gaussian filtering |
| KR101426298B1 (en) * | 2014-01-24 | 2014-08-13 | (주)인펙비전 | apparatus and method for compensating image for enhancing fog removing efficiency |
| 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 |
| US20160071244A1 (en) * | 2014-09-04 | 2016-03-10 | National Taipei University Of Technology | Method and system for image haze removal based on hybrid dark channel prior |
| CN112465708A (en) * | 2020-10-23 | 2021-03-09 | 南京理工大学 | Improved image defogging method based on dark channel |
| CN113487509A (en) * | 2021-07-14 | 2021-10-08 | 杭州电子科技大学 | Remote sensing image fog removing method based on pixel clustering and transmissivity fusion |
| CN114119411A (en) * | 2021-11-24 | 2022-03-01 | 湖南中科助英智能科技研究院有限公司 | Fog noise video image recovery method, device, equipment and medium |
| CN114219732A (en) * | 2021-12-15 | 2022-03-22 | 大连海事大学 | Image defogging method and system based on sky region segmentation and transmissivity refinement |
Non-Patent Citations (4)
| Title |
|---|
| HANG YU: "A Novel Nighttime Dehazing Model Integrating Retinex Algorithm and Atmospheric Scattering Model", 2022 3RD INTERNATIONAL CONFERENCE ON GEOLOGY, MAPPING AND REMOTE SENSING (ICGMRS) * |
| 张晨: "基于融合与高斯加权暗通道的单幅图像去雾算法", 光子学报, pages 1 - 6 * |
| 陈长华;刘煜;崔强;: "结合饱和度运算和暗通道理论的遥感图像去雾算法", 计算机工程与应用, no. 05 * |
| 马文君;刘金虎;王小鹏;孙士伟;: "结合Lab空间和单尺度Retinex的自适应图像去雾算法", 应用光学, no. 01 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115861104B (en) | 2023-10-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107301623B (en) | Traffic image defogging method and system based on dark channel and image segmentation | |
| Gao et al. | Sand-dust image restoration based on reversing the blue channel prior | |
| CN107301624B (en) | Convolutional neural network defogging method based on region division and dense fog pretreatment | |
| CN111598791B (en) | Image defogging method based on improved dynamic atmospheric scattering coefficient function | |
| CN110782407B (en) | Single image defogging method based on sky region probability segmentation | |
| CN107798670B (en) | Dark channel prior image defogging method using image guide filter | |
| CN108133462B (en) | Single image restoration method based on gradient field region segmentation | |
| CN108805826B (en) | Method for improving defogging effect | |
| CN114219732A (en) | Image defogging method and system based on sky region segmentation and transmissivity refinement | |
| CN105023246B (en) | A kind of image enchancing method based on contrast and structural similarity | |
| CN112750089B (en) | Optical remote sensing image defogging method based on local block maximum and minimum pixel prior | |
| CN108765316B (en) | Mist concentration self-adaptive judgment method | |
| CN112419163A (en) | Single image weak supervision defogging method based on priori knowledge and deep learning | |
| CN109360169B (en) | Signal processing method for removing rain and mist of single image | |
| CN109191405B (en) | Aerial image defogging algorithm based on transmittance global estimation | |
| CN109241865B (en) | Vehicle detection segmentation algorithm under weak contrast traffic scene | |
| CN115619662A (en) | Image defogging method based on dark channel prior | |
| CN115861104B (en) | Remote sensing image defogging method based on transmissivity refinement | |
| CN115170437A (en) | Fire scene low-quality image recovery method for rescue robot | |
| CN112598777B (en) | Haze fusion method based on dark channel prior | |
| CN112907461B (en) | Defogging enhancement method for infrared foggy-day degraded image | |
| CN112949389A (en) | Haze image target detection method based on improved target detection network | |
| CN115496694B (en) | Method for recovering and enhancing underwater image based on improved image forming model | |
| Han et al. | Single image dehazing method via sky-regions segmentation and dark channel prior | |
| CN113160073B (en) | Remote sensing image haze removal method combining rolling deep learning and Retinex theory |
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 |