CN104881879B - A kind of remote sensing images haze emulation mode based on dark channel prior - Google Patents
A kind of remote sensing images haze emulation mode based on dark channel prior Download PDFInfo
- Publication number
- CN104881879B CN104881879B CN201510329446.3A CN201510329446A CN104881879B CN 104881879 B CN104881879 B CN 104881879B CN 201510329446 A CN201510329446 A CN 201510329446A CN 104881879 B CN104881879 B CN 104881879B
- Authority
- CN
- China
- Prior art keywords
- mrow
- haze
- image
- remote sensing
- msub
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/536—Depth or shape recovery from perspective effects, e.g. by using vanishing points
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30192—Weather; Meteorology
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
A kind of remote sensing images haze emulation mode based on dark channel prior, comprises the following steps：1st, remote sensing image data is obtained；2nd, the dark channel diagram of image is calculated；3rd, global atmosphere light estimation；4th, transmission plot is extracted；5th, the emulation of haze remote sensing images；By emulating haze remote sensing images, haze concentration scale can be made, it is known that need not be classified by Artificial Cognition, so that haze true value is provided for image quality evaluation and defogging algorithm research, it is easy and effective, it is time saving and energy saving.
Description
(1) technical field：
The present invention relates to a kind of remote sensing images haze emulation mode based on dark channel prior, belong to the technology of image procossing
Field.
(2) background technology：
Satellite remote sensing is the Important Platform of Information Network, and the important of spatial information is obtained and handle in real time as national
Infrastructure, is the key means of the major areas Scene acquisition of information such as security monitoring, emergency management and rescue, military combat.At present,
Remote sensing technology is widely used to the fields such as environment, the hydrology, meteorology, geology, military affairs.With the geometry level increase of remote sensing image,
The importance of its quality evaluation is also growing.According to the different quality of remote sensing images, select different in-orbit processing mode and
The priority level passed down, can effectively improve in-orbit processing and the real-time Transmission ability of remotely-sensed data, while being also beneficial to improve distant
Feel the accuracy of data automatic interpretation.In the various factors of influence Remote Sensing Image Quality, haze occlusion issue drastically influence
Interpretation of the human eye to view data, while also have impact on the automatic interpretation of remotely-sensed data.The present invention is directed to the haze of remote sensing images
Occlusion issue is studied, the research of the remote sensing images with haze is mostly concentrated at this stage it is how effective go demisting,
And rarely have to the evaluation problem of haze degree and refer to.In the acquisition of remote sensing images, because no fog free images are referred to,
The checking of various defogging algorithms is completed generally by eye-observation.And for the research in terms of haze degree evaluation be even more by
It is trapped in the actual value without mistiness degree.
The present invention dark channel prior for image defogging field is introduced into haze emulation system, first by the greasy weather into
As model and dark channel prior extract the transmission plot of haze image, then based on the transmission plot, appropriate adjusting parameter is substituted into
The remote sensing images with different degrees of haze can be produced into greasy weather imaging model.By the present invention simulate Lai haze figure
Picture, it is visually very much like with actual haze image, and there is provided the true value of haze, can be not only used for the effective of defogging algorithm
Property assess, it can also be used to the research and checking of mistiness degree evaluation method.
(3) content of the invention：
1st, purpose：It is an object of the invention to provide a kind of remote sensing images haze emulation mode based on dark channel prior,
Removed for remote sensing images haze and haze degree evaluation provides and refers to true value.This method can emulate obtain different distributions and
Different degrees of haze remote sensing images, the quality and the quality of defogging algorithm of the quality of evaluation evaluation method that can be quantified,
Vital effect is played in haze image quality evaluation and the research of defogging method.
2nd, technical scheme：The present invention is achieved through the following technical solutions.
Description the greasy weather imaging model be：
I (x)=J (x) t (x)+A (1-t (x))
Wherein, I (x) is the image obtained, is haze image, and J (x) is the corresponding picture rich in detail of haze image, and A is global
Atmosphere light, t (x) describes to enter the unscattered light of camera, as transmission plot, and x represents the position in the picture of pixel
Put.In image defogging method, it is known that I (x), global atmosphere light A and transmission plot t (x) are estimated by priori, then to mist
Its imaging model is derived by：
Finally by I (x), A and t (x) are substituted into and be can obtain the image after defogging.In contrast, haze image, which is emulated, is then
Known picture rich in detail, by adjusting atmosphere light and estimation transmission plot, is then updated to greasy weather imaging model to obtain different points
Cloth and different degrees of haze image.Therefore, in haze image emulation, key is to estimate transmission plot, and transmission plot is estimated
Counting then needs first to estimate global atmosphere light, and the present invention is a kind of remote sensing images haze emulation side based on dark channel prior
Method, this method includes such as next step：
Step 1：Remote sensing image data is obtained
In this step, remote sensing images are intercepted from Google Earth (Google Earth).Concrete operations are as follows：
1. Google Earth software is opened, screenshot area is selected；
2. file → preservation → preservation image is clicked, you can obtain remote sensing images.
Some pictures rich in detail and haze image are intercepted by aforesaid operations in Google Earth, wherein haze image includes difference
The mist of distribution and different degrees of mist, picture size are unified, and in the present invention, picture size is 600 × 500.
Step 2：Calculate the dark channel diagram of image
For the remote sensing images of tri- passages of R, G, B, dark channel diagram computational methods are as follows：
1. for each pixel, the value of its tri- passages of R, G, B is compared, using minimum value as the value at the pixel,
Gray-scale map is obtained after being operated to entire image；
2. in this gray-scale map, with the sliding window of one 15 × 15 as regional area from image left to bottom right
Traveled through, calculate the minimum value in the region, the value using the value as whole region just can obtain dark channel diagram.
Step 3：Global atmosphere light A estimation
According to dark channel prior, each regional area there will be likely some the brightness value of at least one Color Channel very
It is low or even close to 0.Preceding 300 pixels for taking gray value maximum in dark channel diagram, generally image total pixel number
0.1%, in the corresponding artwork of these pixels is I (x) position, the maximum of R passages, G passages and channel B is picked out respectively
Value, then takes the average of these three values to be denoted as A0, then global atmosphere light A can be estimated by following formula：
Step 4：Transmission plot is extracted
1. transmission plot rough estimate.It is that each regional area there will be likely some at least one using dark channel prior
The brightness value of Color Channel is very low or even close to 0, and transmission plot estimation formulas is derived in conjunction with greasy weather imaging model：
In formula, IcAny passage in haze image tri- passages of R, G, B is represented, Ω (x) is regional area.According to the public affairs
Formula, substitutes into the global atmosphere light having been estimated that and known haze image I, you can try to achieve rough transmission plot；
2. transmission plot becomes more meticulous.The transmission plot 1. estimated by step has blocking effect, makes transmission to remove blocking effect
Figure is finer, and in the present invention we used guiding filtering, its filtering core is：
Wherein I schemes for guiding, can be gray-scale map, can also use coloured image, ωkRepresent the window centered on pixel k
Mouthful, | ω | represent the pixel quantity in the window, μkWithRespectively navigational figure I is in window ωkInterior average and variance, ε
For adjusting parameter.In the present invention, guiding figure I is the image that the gray value without any information is 0, in order to avoid original
The influence of atural object texture in image.Use Wij(I) transmission plot roughly estimated is filtered, then after filtering after transmission plot
There is no blocking effect, while the distribution of haze can be reflected again.
Step 5：The emulation of haze remote sensing images
Transmission plot just can carry out haze image emulation after extracting according to greasy weather imaging model, i.e., from actual band mist image
Transmission plot t is estimated, clearly image J together substitutes into greasy weather imaging model with a width by it, and adjusts global atmosphere light A, i.e.,
Haze can be added on clear figure, obtain haze analogous diagram.Haze in remote sensing images has different distribution and different dense
Degree, therefore can be divided into two classes in the image of emulation, select according to the actual requirements.Specific implementation process is as follows：
(1) the haze image emulation of same distribution various concentrations
In greasy weather imaging model, picture rich in detail J is, it is known that transmission plot is to estimate to obtain from existing haze remote sensing images
, it determines the substantially distribution of haze in emulating image, is changeless in this step.Global atmosphere light A is then determined
The concentration of haze, is in this step variable in emulating image, and span is [180,255].By adjusting global air
The size of light A values is so as to obtain the image with various concentrations mist.
(2) the haze image emulation of different distributions
In transmission plot extraction step, different transmission plots can be extracted from existing multiple haze remote sensing images,
Because transmission plot reflects the distribution situation of haze, therefore in this step, only same width picture rich in detail need to be fixed, and by difference
Transmission plot be updated in greasy weather imaging model, you can in the clear figure of same width add different distributions mist.It should be noted
It is that global atmosphere light A can be set to fixed value herein, can also be adjusted according to demand, and general A takes 220.
3rd, advantage and effect
It is an advantage of the invention that：By emulating haze remote sensing images, haze concentration scale can be made, it is known that without by people
Work identification is classified, so that haze true value is provided for image quality evaluation and defogging algorithm research, easy and effective, time saving province
Power.
According to different demands, by adjust overall situation atmosphere light A and transmission plot t can simulate it is various have different distributions with
The haze remote sensing images of various concentrations, these emulation haze images are visually very much like with true haze image, can be used for
The researchs such as image quality evaluation, assessment defogging algorithm performance.
(4) illustrate
The flow chart of Fig. 1 emulation modes of the present invention.
(5) embodiment
Embodiments of the present invention are made further by technical scheme for a better understanding of the present invention below in conjunction with accompanying drawing
Description：
The present invention realizes that flow chart of the invention is as shown in Figure 1 under MATLAB 2013a programmed environments.Allocation of computer
Using：Intel Core i5-2400 processors, dominant frequency 3.1GHz, internal memory 2GB, operating system are 32-bit Windows 7.This hair
Bright is a kind of remote sensing images haze emulation mode based on dark channel prior, specifically includes following steps：
Step 1：Haze remote sensing images of the interception from Google Earth are read under MATLAB 2013a environment；
In this step, remote sensing images are intercepted from Google Earth (Google Earth).Concrete operations are as follows：
1. Google Earth software is opened, screenshot area is selected；
2. file → preservation → preservation image is clicked, you can obtain remote sensing images.
Some pictures rich in detail and haze image are intercepted by aforesaid operations in Google Earth, wherein haze image includes difference
The mist of distribution and different degrees of mist, picture size are unified, and in the present invention, picture size is 600 × 500.
Step 2：Calculate the dark channel diagram of image
For the remote sensing images of tri- passages of R, G, B, dark channel diagram computational methods are as follows：
1. for each pixel, the value of its tri- passages of R, G, B is compared, using minimum value as the value at the pixel,
Gray-scale map is obtained after being operated to entire image；
2. in this gray-scale map, with the sliding window of one 15 × 15 as regional area from image left to bottom right
Traveled through, calculate the minimum value in the region, the value using the value as whole region just can obtain dark channel diagram；
Step 3：Global atmosphere light A estimation
First 300 (the 0.1% of total pixel number) pixels for taking gray value maximum in dark channel diagram, in these pixels correspondence
Artwork be I (x) position, the maximum of R passages, G passages and channel B is picked out respectively, the equal of these three values is then taken
Value is denoted as A0, then global atmosphere light can be estimated by following formula：
Step 4：Transmission plot is extracted
1. transmission plot rough estimate.Transmission plot estimation formulas is derived using dark channel prior and greasy weather imaging model：
According to the formula, the dark channel diagram for substituting into the global atmosphere light having been estimated that and having calculated can be tried to achieve roughly
Transmission plot；
2. transmission plot becomes more meticulous.The transmission plot 1. estimated by step has blocking effect, makes transmission to remove blocking effect
Figure is finer, and in the present invention we used guiding filtering, its filtering core is：
Wherein guiding figure I uses the image that the gray value without any information is 0, in order to avoid original image
The influence of middle atural object texture.Use Wij(I) transmission plot roughly estimated is filtered, filter radius is set to 20, adjusting parameter
It is set to 10-3, so after filtering after transmission plot there is no blocking effect, while the distribution of haze can be reflected again.
Step 5：The emulation of haze remote sensing images
Transmission plot just can carry out haze image emulation after extracting according to greasy weather imaging model, i.e., from actual band mist image
Transmission plot t is estimated, clearly image J together substitutes into greasy weather imaging model with a width by it, and adjusts global atmosphere light A, i.e.,
Haze can be added on clear figure, obtain haze analogous diagram.Haze in remote sensing images has different distribution and different dense
Degree, therefore can be divided into two classes in the image of emulation, select according to the actual requirements.Specific implementation process is as follows：
(1) the haze image emulation of same distribution various concentrations
In greasy weather imaging model, picture rich in detail J is, it is known that transmission plot is to estimate to obtain from existing haze remote sensing images
, it determines the substantially distribution of haze in emulating image, is changeless in this step, global atmosphere light A is then determined
The concentration of haze, is in this step variable in emulating image, and span is [180,255].By adjusting global air
The size of light A values is so as to obtain the image with various concentrations mist.
(2) the haze image emulation of different distributions
In transmission plot extraction step, different transmission plots can be extracted from existing multiple haze remote sensing images,
Because transmission plot reflects the distribution situation of haze, therefore in this step, only same width picture rich in detail need to be fixed, and by difference
Transmission plot be updated in greasy weather imaging model, you can in the clear figure of same width add different distributions mist.It should be noted
It is that global atmosphere light A can be set to fixed value herein, can also be adjusted according to demand, and general A takes 220.
Claims (1)
1. a kind of remote sensing images haze emulation mode based on dark channel prior, it is characterised in that：This method includes following several
Step：
Step 1：Remote sensing image data is obtained
In this step, remote sensing images are intercepted from Google Earth, and concrete operations are as follows：
1. Google Earth software is opened, screenshot area is selected；
2. file → preservation → preservation image is clicked, that is, obtains remote sensing images；
Picture rich in detail and haze image are intercepted by aforesaid operations in Google Earth, wherein haze image includes the mist of different distributions
And different degrees of mist, picture size unification, picture size is 600 × 500；
Step 2：Calculate the dark channel diagram of image
For the remote sensing images of tri- passages of R, G, B, dark channel diagram computational methods are as follows：
1. for each pixel, the value of its tri- passages of R, G, B is compared, using minimum value as the value at the pixel, to whole
Gray-scale map is obtained after width image manipulation；
2. in this gray-scale map, with the sliding window of one 15 × 15 as regional area carrying out left to bottom right from image
Traversal, calculates the minimum value in the region, the value using the value as whole region just obtains dark channel diagram；
Step 3：Global atmosphere light A estimation
According to dark channel prior, each regional area exist at least one Color Channel brightness value it is very low in addition close to
0, preceding 300 pixels for taking gray value maximum in dark channel diagram, are the 0.1% of image total pixel number, in these pixels correspondence
Artwork be I (x) position, the maximum of R passages, G passages and channel B is picked out respectively, the equal of these three values is then taken
Value is denoted as A0, then global atmosphere light A is estimated by following formula：
<mrow>
<mi>A</mi>
<mo>=</mo>
<msub>
<mi>A</mi>
<mn>0</mn>
</msub>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
Step 4：Transmission plot is extracted
1. transmission plot rough estimate：It is that each regional area has the bright of at least one Color Channel using dark channel prior
Angle value is very low or even close to 0, and transmission plot estimation formulas is derived in conjunction with greasy weather imaging model：
<mrow>
<mover>
<mi>t</mi>
<mo>~</mo>
</mover>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mn>1</mn>
<mo>-</mo>
<mfrac>
<mrow>
<munder>
<mrow>
<mi>m</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
<mrow>
<mi>y</mi>
<mo>&Element;</mo>
<mi>&Omega;</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
</mrow>
</munder>
<mrow>
<mo>(</mo>
<munder>
<mrow>
<mi>m</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
<mrow>
<mi>c</mi>
<mo>&Element;</mo>
<mo>{</mo>
<mi>r</mi>
<mo>,</mo>
<mi>g</mi>
<mo>,</mo>
<mi>b</mi>
<mo>}</mo>
</mrow>
</munder>
<msup>
<mi>I</mi>
<mi>c</mi>
</msup>
<mo>(</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>)</mo>
</mrow>
<msub>
<mi>A</mi>
<mn>0</mn>
</msub>
</mfrac>
</mrow>
In formula, IcAny passage in haze image tri- passages of R, G, B is represented, Ω (x) is regional area；According to the formula, generation
Enter the global atmosphere light A and known haze image I having been estimated that, that is, try to achieve rough transmission plot；
2. transmission plot becomes more meticulous：The transmission plot 1. estimated by step has blocking effect, makes transmission plot more to remove blocking effect
Plus it is fine, we used guiding filtering, its filtering core is：
<mrow>
<msub>
<mi>W</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>I</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mrow>
<mo>|</mo>
<mi>&omega;</mi>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>:</mo>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>&Element;</mo>
<msub>
<mi>&omega;</mi>
<mi>k</mi>
</msub>
</mrow>
</munder>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>+</mo>
<mfrac>
<mrow>
<mo>(</mo>
<msub>
<mi>I</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>&mu;</mi>
<mi>k</mi>
</msub>
<mo>)</mo>
<mo>(</mo>
<msub>
<mi>I</mi>
<mi>j</mi>
</msub>
<mo>-</mo>
<msub>
<mi>&mu;</mi>
<mi>k</mi>
</msub>
<mo>)</mo>
</mrow>
<mrow>
<msubsup>
<mi>&sigma;</mi>
<mi>k</mi>
<mn>2</mn>
</msubsup>
<mo>+</mo>
<mi>&epsiv;</mi>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
</mrow>
Wherein I schemes for guiding, and it is gray-scale map, can also use coloured image, ωkThe window centered on pixel k is represented, | ω | table
Show the pixel quantity in the window, μ k andRespectively navigational figure I is in window ωkInterior average and variance, ε joins for adjustment
Number；
Step 5：The emulation of haze remote sensing images
Transmission plot just carries out haze image emulation after extracting according to greasy weather imaging model, i.e., estimated from actual band mist image
Transmission plot t, by it, clearly image J together substitutes into greasy weather imaging model with a width, and adjusts global atmosphere light A, can be clear
Haze is added on clear figure, haze analogous diagram is obtained；Haze in remote sensing images has different distributions and different concentration, therefore
It is divided into two classes in the image of emulation, selects according to the actual requirements, specific implementation process is as follows：
(1) the haze image emulation of same distribution various concentrations
In greasy weather imaging model, picture rich in detail J, it is known that transmission plot is estimated to obtain from existing haze remote sensing images, it
The substantially distribution of haze in emulating image is determined, is changeless in this step；Global atmosphere light A then determines emulation
The concentration of haze in image, is in this step variable, and span is [180,255], by adjusting global atmosphere light A values
Size is so as to obtain the image with various concentrations mist；
(2) the haze image emulation of different distributions
In transmission plot extraction step, different transmission plots are extracted from existing multiple haze remote sensing images, due to transmission
Figure reflects the distribution situation of haze, therefore in this step, need to only fix same width picture rich in detail, and by different transmission plots
It is updated in greasy weather imaging model, the mist of different distributions can be added in the clear figure of same width；It should be noted that global big
Gas light A is set to fixed value herein, can also be adjusted according to demand, A takes 220.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510329446.3A CN104881879B (en) | 2015-06-15 | 2015-06-15 | A kind of remote sensing images haze emulation mode based on dark channel prior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510329446.3A CN104881879B (en) | 2015-06-15 | 2015-06-15 | A kind of remote sensing images haze emulation mode based on dark channel prior |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104881879A CN104881879A (en) | 2015-09-02 |
| CN104881879B true CN104881879B (en) | 2017-10-13 |
Family
ID=53949363
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510329446.3A Active CN104881879B (en) | 2015-06-15 | 2015-06-15 | A kind of remote sensing images haze emulation mode based on dark channel prior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104881879B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809647B (en) * | 2016-03-31 | 2020-06-05 | 北京奇虎科技有限公司 | Automatic defogging photographing method, device and equipment |
| CN105913385B (en) * | 2016-03-31 | 2018-12-25 | 宇龙计算机通信科技(深圳)有限公司 | Clarification method, system and the image processing apparatus of haze image |
| CN105976337B (en) * | 2016-05-10 | 2018-12-18 | 长安大学 | A kind of image defogging method based on intermediate value guiding filtering |
| CN106169176A (en) * | 2016-06-27 | 2016-11-30 | 上海集成电路研发中心有限公司 | A kind of image defogging method |
| CN106446957B (en) * | 2016-10-08 | 2019-02-22 | 常熟理工学院 | A kind of haze image classification method based on random forest |
| CN107103591B (en) * | 2017-03-27 | 2020-01-07 | 湖南大学 | Single image defogging method based on image haze concentration estimation |
| CN107578386A (en) * | 2017-09-01 | 2018-01-12 | 荆门程远电子科技有限公司 | A kind of optimization defogging processing method of unmanned plane shooting image |
| CN108681990B (en) * | 2018-04-04 | 2022-05-24 | 高明合 | Real-time haze early warning method and system |
| CN111504271A (en) * | 2020-04-16 | 2020-08-07 | 中国科学院光电技术研究所 | Visible light image detector with parameters capable of being bound |
| CN112750089B (en) * | 2020-12-27 | 2023-03-28 | 同济大学 | Optical remote sensing image defogging method based on local block maximum and minimum pixel prior |
| CN116188633B (en) * | 2023-04-25 | 2023-07-18 | 之江实验室 | Method, device, medium and electronic equipment for generating simulated remote sensing image |
| CN117351426A (en) * | 2023-10-24 | 2024-01-05 | 秦皇岛燕大滨沅科技发展有限公司 | Bulk cargo port dust monitoring method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102654914A (en) * | 2011-03-04 | 2012-09-05 | 富士通株式会社 | Method for accelerating image haze removal by utilizing image processing unit |
| CN102663694A (en) * | 2012-03-30 | 2012-09-12 | 大连理工大学 | Digital fog effect filter method based on dark primary color channel prior principle |
| CN103218778A (en) * | 2013-03-22 | 2013-07-24 | 华为技术有限公司 | Image and video processing method and device |
| CN104408757A (en) * | 2014-11-07 | 2015-03-11 | 吉林大学 | Method and system for adding haze effect to driving scene video |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101568971B1 (en) * | 2011-08-03 | 2015-11-13 | 인디안 인스티튜트 오브 테크놀로지, 카라그푸르 | Method and system for removal of fog, mist or haze from images and videos |
-
2015
- 2015-06-15 CN CN201510329446.3A patent/CN104881879B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102654914A (en) * | 2011-03-04 | 2012-09-05 | 富士通株式会社 | Method for accelerating image haze removal by utilizing image processing unit |
| CN102663694A (en) * | 2012-03-30 | 2012-09-12 | 大连理工大学 | Digital fog effect filter method based on dark primary color channel prior principle |
| CN103218778A (en) * | 2013-03-22 | 2013-07-24 | 华为技术有限公司 | Image and video processing method and device |
| CN104408757A (en) * | 2014-11-07 | 2015-03-11 | 吉林大学 | Method and system for adding haze effect to driving scene video |
Non-Patent Citations (2)
| Title |
|---|
| 《Dehazing for Image and Video Using Guided Filter》;Z Lin等;《Open Journal of Applied Sciences》;20120131;第2卷(第4期);123-127 * |
| 《一种基于SVM分类的雾图自动检测方法》;胡众义等;《计算机仿真》;20150228;第32卷(第02期);342-346 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104881879A (en) | 2015-09-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104881879B (en) | A kind of remote sensing images haze emulation mode based on dark channel prior | |
| CN108830150B (en) | One kind being based on 3 D human body Attitude estimation method and device | |
| CN104299192B (en) | A kind of single image to the fog method based on atmospheric light scattering physical model | |
| CN102750674B (en) | Video image defogging method based on self-adapting allowance | |
| CN109559310A (en) | Power transmission and transformation inspection image quality evaluating method and system based on conspicuousness detection | |
| CN103914813B (en) | The restored method of colored haze image defogging and illumination compensation | |
| CN102231791B (en) | Video image defogging method based on image brightness stratification | |
| CN110378865A (en) | A kind of greasy weather visibility intelligence hierarchical identification method and system under complex background | |
| CN103268470B (en) | Object video real-time statistical method based on any scene | |
| CN104200445A (en) | Image defogging method with optimal contrast ratio and minimal information loss | |
| CN105608417B (en) | Traffic lights detection method and device | |
| CN106940882A (en) | A kind of transformer substation video image clarification method for meeting human-eye visual characteristic | |
| CN109035188A (en) | A kind of intelligent image fusion method based on target signature driving | |
| CN103077504B (en) | A kind of image defogging method capable based on self-adaptation illumination calculation | |
| CN102663694A (en) | Digital fog effect filter method based on dark primary color channel prior principle | |
| CN106530266B (en) | A kind of infrared and visible light image fusion method based on region rarefaction representation | |
| CN104463869B (en) | A kind of video flame image composite identification method | |
| CN109376676A (en) | Highway engineering site operation personnel safety method for early warning based on unmanned aerial vehicle platform | |
| CN106709901A (en) | Simulation fog image generation method based on depth priori | |
| CN106815826A (en) | Night vision image Color Fusion based on scene Recognition | |
| CN108022225A (en) | Based on the improved dark channel prior image defogging algorithm of quick Steerable filter | |
| DE202014010866U1 (en) | Specifiability of images | |
| CN106709903A (en) | PM2.5 concentration prediction method based on image quality | |
| CN110223240A (en) | Image defogging method, system and storage medium based on color decaying priori | |
| CN105631825B (en) | Based on the image defogging method for rolling guiding |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |