CN108280810A - Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image - Google Patents
Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image Download PDFInfo
- Publication number
- CN108280810A CN108280810A CN201810018545.3A CN201810018545A CN108280810A CN 108280810 A CN108280810 A CN 108280810A CN 201810018545 A CN201810018545 A CN 201810018545A CN 108280810 A CN108280810 A CN 108280810A
- Authority
- CN
- China
- Prior art keywords
- area
- cloud
- spissatus
- coverage
- covered areas
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 9
- 230000003287 optical effect Effects 0.000 title abstract 4
- 238000000034 method Methods 0.000 claims abstract description 37
- 239000011159 matrix material Substances 0.000 claims abstract description 14
- 238000012216 screening Methods 0.000 claims abstract description 13
- 230000008439 repair process Effects 0.000 claims abstract description 8
- 238000000605 extraction Methods 0.000 claims abstract description 5
- VMXUWOKSQNHOCA-UKTHLTGXSA-N ranitidine Chemical compound [O-][N+](=O)\C=C(/NC)NCCSCC1=CC=C(CN(C)C)O1 VMXUWOKSQNHOCA-UKTHLTGXSA-N 0.000 claims description 13
- 230000002123 temporal effect Effects 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 5
- 238000004040 coloring Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000000007 visual effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 239000003595 mist Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000007794 visualization technique Methods 0.000 description 1
Classifications
-
- G06T5/70—
-
- G06T5/77—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
Abstract
The invention provides an automatic processing method for repairing a cloud coverage area of a single-time phase optical remote sensing image, and aims at the limitation of a traditional cloud coverage area image repairing method in an optical remote sensing image. Firstly, in a thick cloud coverage area extraction stage, providing an extraction method based on color and texture characteristics, and screening a thick cloud coverage area by utilizing RGB color space and co-occurrence matrix contrast; secondly, in a cloud coverage area repairing stage, an improved Criminisi algorithm is provided for repairing a thick cloud coverage area and a thin cloud coverage area. According to the method, the cloud coverage area can be efficiently repaired without multi-temporal data, historical data and multi-source data, and the rapid and reliable emergency application of the optical remote sensing image in the repair of the cloud coverage area is realized.
Description
Technical field
The present invention relates to a kind of processing methods of remote sensing images, are covered in particular for Mono temporal remote sensing image medium cloud
Effective reparation in region and visualization method for improving.
Background technology
Remote sensing image is the significant data source in the remote sensing applications such as land use, meteorology, environmental monitoring field.But optics
Image is easier to be influenced by cloud and mist, to subsequently bringing larger difficulty, such as remote sensing images to the analysis of image and interpretation
The identification and classification of middle atural object；In addition also there is larger impact to the visuality of remote sensing images.Therefore, with image processing techniques,
Removal weakens influence of the cloud layer in remote sensing images application, improve the availability of remotely-sensed data and interpretability have it is very heavy
Want Practical significance.Traditional remote sensing images go cloud restorative procedure that can be broadly divided into three categories：Based on prior model, it is based on same district
The not homologous remotely-sensed data in domain and be based on homologous multi-temporal remote sensing data.Wherein, the cloud covered areas reparation side based on prior model
Method is to be established by the sparse dictionary that a large amount of remote sensing images are carried out with feature mostly, and in repairing phase using established dilute
The reparation that dictionary carries out cloud covered areas default data, such as Northwestern Polytechnical University are dredged, Li Ying etc. is learnt dilute by characteristic block
It dredges dictionary and carries out cloud covered areas reparation, such method needs a large amount of history remotely-sensed data to be modeled.Not based on same region
The cloud covered areas restorative procedure of homologous remotely-sensed data, most of is all to utilize in the same area multispectral image and infrared image not
The image information of cloud covered areas is reconstructed with the information of wave band.Such as, University of Electronic Science and Technology, Wang Yong etc., by visible light and close red
The handling result after cloud is gone in wave section remote sensing images, reparation.Based on the cloud covered areas restorative procedure of homologous multi-temporal remote sensing data,
Mainly the domain containing cloud covered areas of present image is predicted using the cloudless overlay area image of different phases.Such as, the Tianjin Chinese is engraved
Development in science and technology Co., Ltd, Zou Yanhua etc. pass through the method for having cloud and cloudless image rigid registrations to two width difference phases
Cloud covered areas is repaired.
To sum up have method, in terms of remote sensing image cloud covered areas reparation, has method and be required for the same area greatly
The prior information of the not homologous data of multi-temporal data, the same area or historical data.With various emergent accident feelings
The increase of condition occurrence frequency, if needing the support of multidate, multi-source or historical data using above method.However, in crowd
These conditions are unable to get satisfaction in mostly emergent remote sensing data application.Therefore it is badly in need of invention one kind and is based only upon Mono temporal remote sensing images
Analyzing processing, the method that directly can effectively carry out cloud covered areas reparation.
Invention content
For the limitation of above-mentioned traditional remote sensing images cloud covered areas restorative procedure, the present invention proposes a kind of Mono temporal optics
The automatic processing method of remote sensing images cloud covered areas reparation, domestic at present there has been no the reports of such method.
Method provided by the invention is extracted by the spissatus area of coverage based on color and textural characteristics and based on improvement
The cloud covered areas of Criminisi algorithms repairs two key step compositions, is as follows：
The first step is extracted based on the spissatus area of coverage of color and textural characteristics
Have obviously relative to other atural objects for the spissatus area of coverage color of big visual field remote sensing image and textural characteristics
The characteristics of difference, this patent carry out the doubtful spissatus area of coverage the preliminary sieve based on color characteristic first with RGB color space
Choosing；Then, using this parameter of gray level co-occurrence matrixes contrast, the candidate cloud based on textural characteristics is carried out to the spissatus area of coverage and is covered
Cover region confirms.
Second step is based on the cloud covered areas reparation for improving Criminisi algorithms
Based on the spissatus overlay area that previous step filters out, this patent covers cloud using improved Criminisi algorithms
It is repaired in area.Specific steps include 5 steps：Cloud covered areas removes the determination of priority, remote sensing atural object best matching blocks
It determines, rechecked using the cloud covered areas reparation of known match block, thin cloud covered areas and smothing filtering and retain final result.
According to an aspect of the invention, there is provided a kind of automatic place of Mono temporal remote sensing image cloud covered areas reparation
Reason method is used for the emergent application of cloud covered areas reparation, includes the following steps：
First, the stage is extracted in the spissatus area of coverage, proposes a kind of extracting method based on color and textural characteristics, utilized
The colouring information of the spissatus area of coverage in RGB color space, preliminary screening go out the spissatus area of coverage, recycle co-occurrence matrix contrast institute
The texture features of reflection, completion extract the spissatus area of coverage；Then, in cloud covered areas repairing phase, a kind of improvement is proposed
Criminisi algorithms, to repairing for the spissatus area of coverage and thin cloud covered areas, this method without multi-temporal data, be not necessarily to
Historical data can be efficiently completed cloud covered areas reparation without multi-source data, realize the fast and reliable of remote sensing image
The emergent application of cloud covered areas reparation.
Description of the drawings
Fig. 1 is a kind of the automatic of Mono temporal remote sensing image cloud covered areas reparation according to an embodiment of the invention
The flow chart of processing method.
Specific implementation mode
It is described below how method provided by the present invention is embodied.
Fig. 1 is a kind of the automatic of Mono temporal remote sensing image cloud covered areas reparation according to an embodiment of the invention
The flow chart of processing method, this method include：
The first step：It is extracted based on the spissatus area of coverage of color and textural characteristics
This patent according to spissatus overlay area in big visual field remote sensing image RGB color space distribution character, it is right
The doubtful spissatus area of coverage carries out preliminary screening, obtains the candidate spissatus area of coverage.Secondly, the image based on gray level co-occurrence matrixes is extracted
Textural characteristics filter out the candidate region with spissatus area of coverage texture feature to color characteristic and confirm, specific steps packet
It includes：
Spissatus area of coverage preliminary screening of (1.1) step based on colouring information
Since the highlight color feature of cloud covered areas in big visual field remote sensing image is its most significant characteristic attribute, because
This can be using the comprehensive judgement of each component value of RGB color space as the condition of preliminary screening.Image is traversed pixel-by-pixel, takes colour
Components R, G, B triple channel are all higher than 210 pixel value region as the doubtful spissatus area of coverage in image.
Candidate cloud covered areas of (1.2) step based on textural characteristics confirms
Since (1.1) step may include some highlighted ground from the spissatus area of coverage of candidate obtained in whole picture remote sensing images
The interference of object.To reject false-alarm interference, the image texture characteristic based on gray level co-occurrence matrixes is extracted to these candidate regions.In order to
The texture features of candidate regions are quickly and effectively described, mainly using in co-occurrence matrix derived character compared with can intuitively reflect cloud covered areas
The parameter of texture feature -- co-occurrence matrix contrast.That reflects the high frequency detail degree of image and texture rill depth degree,
Generally have the characteristics that seamlessly transit according to the local grain of cloud covered areas, high frequency detail smaller using co-occurrence matrix contrast
Less and this more shallow characteristic of rill, the excessively high non-cloud highlight regions of co-occurrence matrix contrast are excluded.Realize that candidate cloud covered areas is true
Recognize.
Second step：Based on the cloud covered areas reparation for improving Criminisi algorithms
Based on the spissatus overlay area that previous step filters out, cloud covered areas is carried out using improved Criminisi algorithms
It repairs.The algorithm mainly removes the determination of priority by cloud covered areas, the determination of remote sensing atural object best matching blocks, utilizes known
Cloud covered areas reparation, the reinspection of thin cloud covered areas and smothing filtering with block simultaneously retain five part compositions of final result：
(2.1) buyun area of coverage removes the determination of priority
The doubtful spissatus area of coverage obtained for previous step, it is thus necessary to determine that removal sequence, it is therefore an objective to ensure the line in image
Property structure-borne and object boundary be connected to.Wherein, the priority calculation formula of the block centered on the marginal point P of target area is such as
Under：
P (p)=C (p) D (p)
Wherein：C (p) is confidence level item, and D (p) is data item, is defined as follows：
Wherein：|Ψp| it is the area for currently choosing block, α is the image standardization factor, npAt p on the edge of target area
Unit normal vector, IpFor point p isophotes.The point of highest priority is chosen as point to be removed in starting stage C (p)=0.
The determination of (2.2) step remote sensing atural object best matching blocks
According to difference of two squares distance and minimum principle, the best remote sensing atural object Texture Matching block in known region is found, it is public
Formula is as follows：
Ψq=mind (Ψp,Ψq)
Wherein：ΨqBe find in the zone and ΨpMost suitable match block.
(2.3) step utilizes the cloud covered areas reparation of known match block
By the remote sensing atural object best matching blocks obtained in step (2.2) corresponding target area position is copied to as object
It sets, then carries out the boundary of target area and the update of confidence value, target and best matching blocks to be removed after finding successively,
The image that spissatus area of coverage reparation is completed finally can be obtained.
The thin cloud covered areas reinspection of (2.4) step
By above-mentioned steps it is found that having filtered out doubtful spissatus time according to color and textural characteristics in (1.1) and (1.2) step
Constituency realizes effective extraction of the spissatus area of coverage.But in remote sensing image, in addition to the spissatus area of coverage, there is also thin clouds
The area of coverage.Since the characterization image characteristic of remote sensing atural object is different, the presence of some of thin cloud covered areas is with can also influencing remote sensing
The identification of object.Therefore, in order to meet the needs of emergency processing, using thin cloud covered areas reinspection policies, this is asked this patent
Topic.The specific steps are：
2.4.1 the differentiation of doubtful thin cloud covered areas is carried out first.Being free of after taking (2.3) step to complete spissatus area of coverage reparation
Spissatus image takes R, G, B triple channel in its RGB color space to be all higher than 190 pixel values and meet cloud covered areas textural characteristics
Region as thin cloud covering candidate regions (concrete operations with 1,1 and 1,2 steps).
2.4.2 the statistics of area accounting is carried out to doubtful thin cloud covered areas.Calculate thin cloud covered areas area in present image
With the ratio of total image area, as comparing threshold value T.
2.4.3 a point strategy is carried out according to area threshold to handle.If less than threshold value T (the present embodiment value is 10%), recognize
Limited for the influence to vision interpretation, it is the Criminisi algorithm reparations being directly improved only to need local route repair, restorative procedure
(concrete operations are with 2.1 to 2.4 steps)；If more than this threshold value T, then it is assumed that influence to vision interpretation can not objective interpretation, no
Thin cloud reparation is carried out, directly exports the spissatus court verdict of 2.3 steps as final result.
(2.5) step smothing filtering simultaneously retains final result
Due to the image that previous step reparation is completed, some non-linear noise spots for restoring to bring always are will produce, therefore pass through
Gaussian filter removal is selected, visuality is promoted, it is final to obtain the image repaired and completed.
The present invention has the following advantages compared with existing restorative procedure：
(1) it is directed in terms of remote sensing image cloud covered areas reparation, existing method is required for greatly the same area more at present
When complicated precondition, this patent such as the not homologous prior information of data or historical data of phase data, the same area propose
A kind of automatic processing method of Mono temporal remote sensing image cloud covered areas reparation.This method makes full use of the color of cloud covered areas
Feature and textural characteristics are then proposed into the screening for the covering candidate regions that rack based on the cloud covered areas for improving Criminisi algorithms
Restorative procedure carries out cloud covered areas reparation.This patent restorative procedure only needs to handle based on Mono temporal remote Sensing Image Analysis, can be straight
It is connected to effect and carries out cloud covered areas reparation.It can be efficiently completed without multi-temporal data, without historical data, without multi-source data
The emergent application of the fast and reliable cloud covered areas reparation of remote sensing image is realized in cloud covered areas reparation.
(2) candidate regions screening stage is covered in cloud, this patent proposes that the doubtful cloud described based on color and textural characteristics is covered
Cover region screening technique.First with rgb space, R, G, B triple channel pixel decision threshold, the condition as preliminary screening are set.
Then, using in gray level co-occurrence matrixes derived character compared with can intuitively reflect cloud covered areas texture feature co-occurrence matrix contrast this
One parameter excludes non-cloud covered areas and highlights remote sensing atural object, further determines that cloud covered areas to be repaired.This method can be from big visual field
In remote sensing image quickly, reliably choose doubtful cloud covered areas.
(3) candidate regions repairing phase is covered in cloud, this patent proposes to cover based on the cloud for improving Criminisi algorithms candidate
Area's restorative procedure.Priority is removed to the spissatus area of coverage to be determined, suitable removal sequence is set first.Again to remote sensing atural object
Best matching blocks be determined, ensure the accuracy of repairing effect.The spissatus area of coverage is carried out using known match block later
It repairs.Then thin cloud covered areas reinspection is carried out, for there being a small amount of thin cloud covered areas, is repaired, last smothing filtering is simultaneously protected
Stay final result.This example demonstrates that this method can quickly repair doubtful cloud covered areas, the accurate of cloud covered areas reparation is improved
Property.
Claims (3)
1. a kind of automatic processing method of Mono temporal remote sensing image cloud covered areas reparation, for the emergent of cloud covered areas reparation
Using including the following steps：
First, the stage is extracted in the spissatus area of coverage, proposes a kind of extracting method based on color and textural characteristics, utilize RGB coloured silks
The colouring information of the spissatus area of coverage in the colour space, preliminary screening go out the spissatus area of coverage, and co-occurrence matrix contrast is recycled to be reflected
Texture features, completion the spissatus area of coverage is extracted；Then, it in cloud covered areas repairing phase, proposes a kind of improved
Criminisi algorithms, to repairing for the spissatus area of coverage and thin cloud covered areas, this method without multi-temporal data, without going through
History data can be efficiently completed cloud covered areas reparation without multi-source data, realize the fast and reliable cloud of remote sensing image
The emergent application of area of coverage reparation.
2. according to the method described in claim 1, it is characterized in that：
In terms of spissatus area of coverage extraction, this patent carries out the doubtful spissatus area of coverage first with RGB color space to be based on face
The preliminary screening of color characteristic using this parameter of gray level co-occurrence matrixes contrast, carries out the spissatus area of coverage to be based on texture later
The candidate cloud covered areas of feature confirms, is as follows：
A) be based on colouring information, the spissatus area of coverage of preliminary screening, including：
Using the comprehensive judgement of each component value of whole picture remote sensing images RGB color space as the condition of preliminary screening, pixel-by-pixel time
Image is gone through, components R in coloured image, G, B triple channel is taken to be all higher than 210 pixel value region as the doubtful spissatus area of coverage,
B textural characteristics) are based on, confirm candidate cloud covered areas, including：
Based on step A) the doubtful spissatus area of coverage of extraction, using co-occurrence matrix contrast level parameter in co-occurrence matrix derived character,
Generally have the characteristics that seamlessly transit according to the local grain of cloud covered areas, the smaller, high frequency detail using co-occurrence matrix contrast
Less and the more shallow characteristic of rill, the excessively high non-cloud highlight regions of co-occurrence matrix contrast are excluded, confirm candidate cloud covered areas.
3. according to the method described in claim 1, it is characterized in that：
Candidate regions repairing phase is covered in cloud, this patent proposes to cover candidate regions reparation based on the cloud for improving Criminisi algorithms
While ensureing that the spissatus area of coverage is repaired the step of judging thin cloud covered areas lash-up recovering necessity is added, specifically in method
Steps are as follows：
A) determine that cloud covered areas removes priority, including：
For the doubtful spissatus area of coverage of acquisition, removal sequence is determined, to ensure the propagation of the linear structure in image and target side
Boundary is connected to, wherein the priority calculation formula of the block centered on the marginal point P of target area is as follows：
P (p)=C (p) D (p)
Wherein：C (p) is confidence level item, and D (p) is data item, is defined as follows：
Wherein：|Ψp| it is the area for currently choosing block, α is the image standardization factor, npFor the unit at p on the edge of target area
Normal vector, IpFor point p isophotes, the point of highest priority is chosen as point to be removed in starting stage C (p)=0,
B remote sensing atural object best matching blocks) are determined, including：
According to difference of two squares distance and minimum principle, best remote sensing atural object Texture Matching block, public affairs used in known region are found
Formula is as follows：
Ψq=min d (Ψp,Ψq)
Wherein：ΨqBe find in the zone and ΨpMost suitable match block,
C known match block) is utilized, cloud covered areas is repaired, including：
By step B) in the remote sensing atural object best matching blocks that obtain copy to corresponding target area position, then carry out target area
Boundary and confidence value update, target and best matching blocks to be removed after finding successively obtain the spissatus area of coverage and repair
The image completed again,
D thin cloud covered areas reinspection) is carried out, including：
Take step C) the spissatus area of coverage repair after be free of spissatus image, take R, G, B triple channel in its RGB color space big
Generally have the characteristics that seamlessly transit in 190 pixel values and according to the local grain of cloud covered areas, be compared using co-occurrence matrix
The characteristic that smaller, high frequency detail is few and rill is more shallow is spent, the excessively high non-cloud highlight regions of co-occurrence matrix contrast are excluded, as thin
Cloud covers candidate regions,
The statistics that area accounting is carried out to doubtful thin cloud covered areas, calculates thin cloud area coverage and total image area in present image
Ratio, as comparing threshold value T (the present embodiment value be 10%),
A point strategy processing is carried out according to area threshold, if being less than threshold value T, then it is assumed that the influence to vision interpretation is limited, only needs office
Portion is repaired, and restorative procedure is the Criminisi algorithm reparations being directly improved, and concrete operations are with step B1) arrive step B4)；
If more than this threshold value T, then it is assumed that influence to vision interpretation can not objective interpretation, repaired without Bao Yun, directly export step
B3 spissatus court verdict),
E) smothing filtering and retain final result
Visuality is promoted, finally by selecting Gaussian filter to remove noise spot based on the image that previous step reparation is completed
Obtain the image repaired and completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810018545.3A CN108280810B (en) | 2018-01-09 | 2018-01-09 | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810018545.3A CN108280810B (en) | 2018-01-09 | 2018-01-09 | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108280810A true CN108280810A (en) | 2018-07-13 |
| CN108280810B CN108280810B (en) | 2020-08-14 |
Family
ID=62803328
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810018545.3A Active CN108280810B (en) | 2018-01-09 | 2018-01-09 | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108280810B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110111259A (en) * | 2019-05-15 | 2019-08-09 | 电子科技大学 | A kind of multisource image anastomosing method based on regional guidance |
| CN111797744A (en) * | 2020-06-28 | 2020-10-20 | 武汉大学 | Multimode remote sensing image matching method based on co-occurrence filtering algorithm |
| CN112561830A (en) * | 2020-12-23 | 2021-03-26 | 安徽大学 | Endoscope image highlight repair method and device |
| CN113298808A (en) * | 2021-06-22 | 2021-08-24 | 哈尔滨工程大学 | Method for repairing building shielding information in tilt-oriented remote sensing image |
| CN116245757A (en) * | 2023-02-08 | 2023-06-09 | 北京艾尔思时代科技有限公司 | Multi-scene universal remote sensing image cloud restoration method and system for multi-mode data |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080031702A (en) * | 2008-02-22 | 2008-04-10 | 강용균 | Method of cloud masking in remote sensing images |
| CN102902956A (en) * | 2012-09-10 | 2013-01-30 | 中国人民解放军理工大学气象学院 | Ground-based visible cloud image recognition processing method |
| CN103020939A (en) * | 2012-12-18 | 2013-04-03 | 武汉大学 | Method for removing large-area thick clouds for optical remote sensing images through multi-temporal data |
| CN103208001A (en) * | 2013-02-06 | 2013-07-17 | 华南师范大学 | Remote sensing image processing method combined with shape self-adaption neighborhood and texture feature extraction |
| CN103472501A (en) * | 2013-09-06 | 2013-12-25 | 中国气象科学研究院 | Cloud and aerial total cloud amount detection method and system |
| CN104484670A (en) * | 2014-10-24 | 2015-04-01 | 西安电子科技大学 | Remote sensing image cloud detection method based on pseudo color and support vector machine |
| CN104778668A (en) * | 2015-04-16 | 2015-07-15 | 电子科技大学 | Optical remote sensing image thin cloud removal method on basis of visible band spectrum statistical characteristics |
| CN105354865A (en) * | 2015-10-27 | 2016-02-24 | 武汉大学 | Automatic cloud detection method and system for multi-spectral remote sensing satellite image |
| CN105426903A (en) * | 2015-10-27 | 2016-03-23 | 航天恒星科技有限公司 | Cloud determination method and system for remote sensing satellite images |
-
2018
- 2018-01-09 CN CN201810018545.3A patent/CN108280810B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080031702A (en) * | 2008-02-22 | 2008-04-10 | 강용균 | Method of cloud masking in remote sensing images |
| CN102902956A (en) * | 2012-09-10 | 2013-01-30 | 中国人民解放军理工大学气象学院 | Ground-based visible cloud image recognition processing method |
| CN103020939A (en) * | 2012-12-18 | 2013-04-03 | 武汉大学 | Method for removing large-area thick clouds for optical remote sensing images through multi-temporal data |
| CN103208001A (en) * | 2013-02-06 | 2013-07-17 | 华南师范大学 | Remote sensing image processing method combined with shape self-adaption neighborhood and texture feature extraction |
| CN103472501A (en) * | 2013-09-06 | 2013-12-25 | 中国气象科学研究院 | Cloud and aerial total cloud amount detection method and system |
| CN104484670A (en) * | 2014-10-24 | 2015-04-01 | 西安电子科技大学 | Remote sensing image cloud detection method based on pseudo color and support vector machine |
| CN104778668A (en) * | 2015-04-16 | 2015-07-15 | 电子科技大学 | Optical remote sensing image thin cloud removal method on basis of visible band spectrum statistical characteristics |
| CN105354865A (en) * | 2015-10-27 | 2016-02-24 | 武汉大学 | Automatic cloud detection method and system for multi-spectral remote sensing satellite image |
| CN105426903A (en) * | 2015-10-27 | 2016-03-23 | 航天恒星科技有限公司 | Cloud determination method and system for remote sensing satellite images |
Non-Patent Citations (7)
| Title |
|---|
| AIXIA DOU ET AL: "Technique of remote sensing image processing in active faults survey", 《 2007 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM》 * |
| P.M ET AL: "Retrieval of Remote Sensing Images Using Color,Texture and Spectral Features", 《INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE AND TECHNOLOGY》 * |
| 刘晓东: "《数字媒体制作项目教程》", 31 July 2017 * |
| 李林: "基于地基云图的云量检测与云状识别", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 王政: "《遥感图像厚云去除方法研究》", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 赵孟银: "《遥感影像去云方法研究》", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
| 魏欣 等: "《改进的Criminisi图像修复算法》", 《计算机工程与应用》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110111259A (en) * | 2019-05-15 | 2019-08-09 | 电子科技大学 | A kind of multisource image anastomosing method based on regional guidance |
| CN111797744A (en) * | 2020-06-28 | 2020-10-20 | 武汉大学 | Multimode remote sensing image matching method based on co-occurrence filtering algorithm |
| CN111797744B (en) * | 2020-06-28 | 2022-06-10 | 武汉大学 | Multimode remote sensing image matching method based on co-occurrence filtering algorithm |
| CN112561830A (en) * | 2020-12-23 | 2021-03-26 | 安徽大学 | Endoscope image highlight repair method and device |
| CN112561830B (en) * | 2020-12-23 | 2022-11-18 | 安徽大学 | Endoscope image highlight repair method and device |
| CN113298808A (en) * | 2021-06-22 | 2021-08-24 | 哈尔滨工程大学 | Method for repairing building shielding information in tilt-oriented remote sensing image |
| CN113298808B (en) * | 2021-06-22 | 2022-03-18 | 哈尔滨工程大学 | Method for repairing building shielding information in tilt-oriented remote sensing image |
| CN116245757A (en) * | 2023-02-08 | 2023-06-09 | 北京艾尔思时代科技有限公司 | Multi-scene universal remote sensing image cloud restoration method and system for multi-mode data |
| CN116245757B (en) * | 2023-02-08 | 2023-09-19 | 北京艾尔思时代科技有限公司 | Multi-scene universal remote sensing image cloud restoration method and system for multi-mode data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108280810B (en) | 2020-08-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108280810A (en) | Automatic processing method for repairing cloud coverage area of single-time phase optical remote sensing image | |
| CN108665487B (en) | Transformer substation operation object and target positioning method based on infrared and visible light fusion | |
| CN111145161A (en) | Method for processing and identifying pavement crack digital image | |
| CN111062931B (en) | Detection method of spliced and tampered image | |
| CN111339948B (en) | Automatic identification method for newly-added buildings of high-resolution remote sensing images | |
| CN103839267A (en) | Building extracting method based on morphological building indexes | |
| CN108416351B (en) | Method and device for determining road surface disease area and computer equipment | |
| CN105117726A (en) | License plate positioning method based on multi-feature area accumulation | |
| CN111310771B (en) | Road image extraction method, device and equipment of remote sensing image and storage medium | |
| Tiwari et al. | Shadow detection and compensation in aerial images using MATLAB | |
| CN107038690B (en) | Moving shadow removing method based on multi-feature fusion | |
| Fengping et al. | Road extraction using modified dark channel prior and neighborhood FCM in foggy aerial images | |
| CN111539980A (en) | Multi-target tracking method based on visible light | |
| Pal et al. | Visibility enhancement techniques for fog degraded images: a comparative analysis with performance evaluation | |
| Deepa et al. | Improved watershed segmentation for apple fruit grading | |
| Ding et al. | Restoration of single sand-dust image based on style transformation and unsupervised adversarial learning | |
| Pratomo et al. | Parking detection system using background subtraction and HSV color segmentation | |
| CN112686222B (en) | Method and system for detecting ship target by satellite-borne visible light detector | |
| AU2020104270A4 (en) | Thangka image inpainting method combining with domain knowledge | |
| CN114119634A (en) | Automatic building extraction method and system combining vegetation elimination and image feature consistency constraint | |
| Ye et al. | Research on Automatic Highway Extraction Technology Based on Spectral Information of Remote Sensing Images. | |
| CN111402265A (en) | Power equipment infrared thermal image segmentation method based on morphological marking | |
| CN114022473B (en) | Horizon detection method based on infrared image | |
| CN117292252A (en) | Fine identification and light cloud automatic removal method for cloud area of remote sensing image | |
| Wang et al. | Morphological Enhancement for Remote Sensing Image Based on GAN Structuring Elements |
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 |