CN105913033A - Multi-layer cloud and single-layer cloud type integrated classification and identification method in remote sensing image - Google Patents

Multi-layer cloud and single-layer cloud type integrated classification and identification method in remote sensing image Download PDF

Info

Publication number
CN105913033A
CN105913033A CN201610237086.9A CN201610237086A CN105913033A CN 105913033 A CN105913033 A CN 105913033A CN 201610237086 A CN201610237086 A CN 201610237086A CN 105913033 A CN105913033 A CN 105913033A
Authority
CN
China
Prior art keywords
cloud
layer
pixel
classification
remote sensing
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
Application number
CN201610237086.9A
Other languages
Chinese (zh)
Other versions
CN105913033B (en
Inventor
王虹霞
许小剑
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beihang University filed Critical Beihang University
Priority to CN201610237086.9A priority Critical patent/CN105913033B/en
Publication of CN105913033A publication Critical patent/CN105913033A/en
Application granted granted Critical
Publication of CN105913033B publication Critical patent/CN105913033B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/13—Satellite images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Astronomy & Astrophysics (AREA)
  • Remote Sensing (AREA)
  • Multimedia (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a multi-layer cloud and single-layer cloud type integrated classification and identification method in a remote sensing image. The method comprises: step one, carrying out cloud-included pixel element identification; step two, carrying out coarse classification and marking on single-layer and multi-layer clouds; step three, carrying out refined single-layer cloud type classification; and step four, carrying out cloud layer phase-state classification and marking. Therefore, cloud-included pixel element identification and integrated cloud layer classification and identification in a remote sensing image can be completed. According to the invention, a precise cloud layer type identification and classification result can be obtained. Distinguishing of a single-layer cloud pixel element and a multi-layer cloud pixel element can be realized; and a phase state of a multi-layer cloud pixel element and a cloud layer type of a single-layer cloud pixel element can be determined. And thus the pixel elements of the remote sensing image can be identified as thirteen kinds of pixel elements including a cloudless sky (cloud-free pixel element), a multi-layer water cloud, a multi-layer ice cloud, a cirrus cloud, a cirrostratus, a vertical extended cloud, a high cumulus cloud, an altostratus, a nimbostratus, a cumulus cloud, a stratocumulus, a stratus cloud, and a filling value, so that diversified image information can be provided for subsequent processing and application of the remote sensing image.

Description

In remote sensing images multi layer cloud and individual layer cloud-type compressive classification with know method for distinguishing
Technical field
The present invention relates to the technical field of remote sensing images cloud layer identification, be specifically related to multi layer cloud and the individual layer varieties of clouds in a kind of remote sensing images Pattern synthesis classification and knowledge method for distinguishing.
Background technology
The existence of cloud layer is the Important Disturbed Factors affecting spaceborne optical pickocff earth observation imaging performance.The kind of cloud layer Various, spatial distribution and the scattering radiation thereof of dissimilar cloud layer are totally different with transmission characteristic, for the clear each cloud type character of research Thus reject the cloud layer impact on observation system, the identification to cloud type has just become the emphasis of research and difficult point (to see document [1-2]).Additionally, either space-based sensor or ground based sensor observation all show that the existence of multi layer cloud is very universal, and to cloud The characteristic of layer has a significant impact (seeing document [3-8]).
According to the phase of cloud layer, cloud layer can be divided into water cloud, ice cloud and mixing phase cloud；Can be by cloud according to the distribution of cloud layer cloud-top height Layer is roughly classified into a high position, middle position and low level；According to cloud layer form, cloud layer can be divided into the type clouds such as stratus, cumulus and cirrus.Several Whether the cloud layer spatial distribution of kind of common type is as it is shown in figure 1, be that multi layer cloud is overlapping and institute by identifying satellite image pixel Corresponding cloud phase and the type of individual layer layer, it is possible to decrease even eliminate the existence impact on observation system of cloud layer.
In Fig. 1, dotted line frame 1 show individual layer cloud, i.e. only exists a stratus layer, remaining wire frame in vertical direction (Z-direction) Shown in be multi layer cloud, multi layer cloud ply there is the most in vertical direction, the number of plies of different is only overlapping cloud layer.Fig. 1 Being two stratus shown in center 2 and frame 4, wherein, two stratus shown in frame 2 are cumulonimbus, the two-layer shown in frame 4 Cloud is respectively cirrus and nimbostratus from top to bottom；Frame 3 show four stratus, the most respectively cirrostratus, altostratus, layer Cumulus and stratus.
Prior art one related to the present invention: the threshold method of individual layer cloud-type identification
The technical scheme of prior art one is:
Now know method for distinguishing for individual layer cloud-type mainly to be differentiated by threshold value, the such as threshold method of individual layer cloud-type identification, literary composition Offer [1] INSAT international satellite thin clouds as plan (International Satellite Cloud Climatology Project, ISCCP) is i.e. according to cloud Top pressure PtopRelatively cloud layer is carried out identification and classification with two variablees of opticalthicknessτ by fixed threshold.ISCCP cloud layer recognition institute The cloud top pressure used and the threshold value of optical thickness as in figure 2 it is shown, the document gives nine kinds of individual layer cloud-type as shown in Figure 2, The most vertically extension cloud refers to the cloud that span is bigger in vertical direction, if the cumulonimbus in Fig. 1 is i.e. that modal one is hung down DS fuzz.
The shortcoming of prior art one
ISCCP cloud layer recognition method realizes the judgement to cloud type according to threshold decision, but ISCCP cloud layer recognition method is assumed Cloud layer to be determined is all individual layer, have ignored the situation that multi layer cloud overlap exists.For Space borne detection, i.e. can only be to going up most The cloud layer of layer differentiates, and cannot realize overlapping cloud layer phase and the identification of type of lower floor.
Prior art two related to the present invention
The technical scheme of prior art two is:
Document [2] combines Moderate Imaging Spectroradiomete (the Moderate Resolution on Terra satellite or Aqua satellite Imaging Spectroradiometer, MODIS) multi layer cloud mark (Multi-Layer Flag, MLF) in data utilizes threshold By pixel, value method judges that its pixel is individual layer cloud pixel, multi layer cloud pixel or vertical extension cloud pixel to image, Fig. 3 gives Go out its multi layer cloud testing process.In figure, desired parameters is: cloud top pressure Ptop, opticalthicknessτ and multi layer cloud mark FML, by (the cloud layer data product of Terra satellite is MOD, Aqua to the cloud layer data product MOD/MYD06 of MODIS data inversion The cloud layer data product of satellite is MYD) obtain.Wherein, MLF be pixel be whether the trust evaluation mark of multilayer pixel, Value is the integer of 0-9, wherein 0 represents clear sky, and 1 represents individual layer cloud, and 2-9 all represents multi layer cloud, and numerical value is represented as the most greatly The confidence level of multi layer cloud is the highest (seeing document [6,7,9]).
Variable Δ P in Fig. 3 decision boxtopComputing formula is:
ΔP t o p = P s - P t o p P s - P t r o p - - - ( 1 )
P in formulasFor earth's surface pressure, PtopFor cloud top pressure, PtropPressure is pushed up for troposphere.The earth's surface pressure wherein used is ozone layer inspection Measurement equipment (Dutch/Finnish Ozone Monitoring Instrument, OMI) detection data.
The shortcoming of prior art two
Document [2] is although achieving individual layer cloud pixel and the differentiation of multi layer cloud pixel, but still suffers from following deficiency: the method only will Satellite image pixel is identified as individual layer cloud pixel, multi layer cloud pixel and vertical extension cloud pixel three class, has not both been shown in detail in individual layer cloud Cloud type, do not provide the phase of multi layer cloud yet.
Bibliography:
[1] http://isccp.giss.nasa.gov/cloudtypes.html#DIAGRAM, 2016.
[2]J.Joiner,A.P.Vasilkov,P.K.Bhartia,G.Wind,S.Platnick,W.P.Menzel, Detection of Multi-layer and Vertically-extended Clouds Using A-train Sensors.Atmospheric Measurement Techniques,2010.3:pp.233–247.
[3]R.Frey,B.A.Baum,A.Heidinger,S.Ackerman,B.Maddux,P.Menzel,MODIS CTP(MOD06)Webinar#7.2014.
[4]J.D.Spinhirne,et al.,Cloud and Aerosol Measurements from GLAS:Overview and Initial Results.Geophysical Research Letters,2005.32(L22S03):pp.1-5.
[5]A.Behrangi,S.P.F.Casey and B.H.Lambrigtsen,Three-dimensional Distribution of Cloud Types Over the USA and Surrounding Areas Observed by CloudSat.International Journal of Remote Sensing,2012.33(16):pp.4856-4870.
[6]W.P.Menzel,R.A.Frey and B.A.Baum,Cloud Top Properties and Cloud Phase Algorithm Theoretical Basis Document.2015.pp.1-73.
[7]S.Platnick,et al.,MODIS Cloud Optical Properties:User Guide for the Collection 6 Level-2 MOD06/MYD06 Product and Associated Level-3 Datasets.2015.
[8]J.D.Spinhirne,S.P.Palm and W.D.Hart,Antarctica Cloud Cover for October 2003 from GLAS Satellite Lidar Profiling.Geophysical Research Letters,2005.32(22):pp.1-4.
[9] http://modis-atmos.gsfc.nasa.gov/, 2016.
Summary of the invention
The technical problem to be solved is: for the problem in above-mentioned satellite image cloud type identification, the present invention proposes A kind of comprehensive multi layer cloud and individual layer cloud-type recognition methods, multi layer cloud is identified, cloud top pressure and the threshold value of optical thickness and The bright temperature characteristics of different phase cloud layer combines, and synthetic determination image picture elements belongs to multi layer cloud pixel or individual layer cloud pixel, simultaneously Can also judge the phase of multi layer cloud pixel, the cloud type etc. of individual layer cloud pixel, perfect to cloud type in remote sensing images Identify.
The technical solution used in the present invention is: the side of multi layer cloud and individual layer cloud-type compressive classification and identification in a kind of remote sensing images Method, the method comprises the steps:
Step-1: have cloud pixel to differentiate
According to cloud mask data, satellite image pixel is divided into cloud pixel, clear sky pixel and Filling power pixel, this Filling power Pixel is for cannot sentence knowledge pixel, and for there being cloud pixel to complete following cloud layer recognition step；
Step-2: individual layer and multi layer cloud rough sort and mark
Utilize earth's surface pressure Ps, troposphere top pressure Ptrop, cloud top pressure Ptop, opticalthicknessτ and multi layer cloud mark FMLAccording to Following decision condition carries out rough sort to cloud layer:
(1) Δ P is mettop> 0.6 and τ > 12 item judge that this pixel is as vertically extending cloud；
(2) Δ P is mettop>0.6, τ<12 and FML>=2 judge that this pixel is as multi layer cloud pixel；
(3) the pixel then preliminary judgement being all unsatisfactory for (1) (2) is common individual layer cloud pixel；
Step-3: individual layer cloud-type sophisticated category
By the pixel that preliminary judgement in step-2 is common individual layer cloud, utilize cloud top pressure Ptop, opticalthicknessτ parameter is according to ISCCP Individual layer cloud pixel is further subdivided into and vertically extends cloud, cirrostratus, cirrus, altostratus, altocumulus, nimbostratus, long-pending by threshold value Cloud, stratocumulus and the pixel of nine kinds of cloud type of stratus；
Step-4: the classification of cloud layer phase and mark
Step-2 will be judged to the pixel of multi layer cloud, utilize 8.5 mu m wavebands and bright temperature difference D of 11 mu m wavebands8.5-11With 11 mu m wavebands With 12 bright temperature difference D of mu m waveband11-12, according to following condition, multi layer cloud pixel is judged:
If meeting D8.5-11≥D11-12Then judge that pixel is as nabivnoy ice cloud pixel；
If meeting D8.5-11<D11-12Then it is judged to multilayer water cloud pixel；
Thus complete that remote sensing images have the discriminating of cloud pixel and the compressive classification of cloud layer and identification.
What technical solution of the present invention was brought has the beneficial effect that
Compared with technology-1 compare with technology-2, multi layer cloud and individual layer cloud-type total score in remote sensing images proposed by the invention Class and recognition methods can obtain the finest cloud type identification and classification results.Use method proposed by the invention, no But can realize individual layer cloud pixel and the differentiation of multi layer cloud pixel, and can determine that phase and the individual layer cloud pixel of multi layer cloud pixel Cloud type, the most at last remote sensing images pixel be identified as in detail clear sky (cloudless pixel), multilayer water cloud, nabivnoy ice cloud, Cirrus, cirrostratus, vertically extend cloud, altocumulus, altostratus, nimbostratus, cumulus, stratocumulus, stratus and Filling power altogether 13 class pixels, subsequent treatment and application for remote sensing images provide the image information of more horn of plenty.
Accompanying drawing explanation
Fig. 1 is the cloud layer spatial distribution schematic diagram of several frequently seen type；
Fig. 2 is cloud top pressure and the optical thickness threshold value (seeing document [1]) of ISCCP cloud layer classification；
Fig. 3 is multi layer cloud and extension cloud detection flow process (seeing document [2])；
Fig. 4 is multi layer cloud and individual layer cloud-type compressive classification and identification process；
Fig. 5 is 2015152.0230 5 minutes band RGB image of MODIS Terra satellite；
Fig. 6 is that MOD06 cloud mask differentiates result；
Fig. 7 is the result of multi layer cloud proposed by the invention and individual layer cloud-type compressive classification and recognition methods, wherein, Fig. 7 (a) Being the judged result of individual layer cloud and multi layer cloud for MODIS multi layer cloud mark MLF, Fig. 7 (b), Fig. 7 (c) is ISCCP Threshold value differentiates result, and Fig. 7 (d) is integrated multilayer cloud and individual layer cloud-type recognition result.Designation explanation corresponding to colour code For: St-stratus, Sc-stratocumulus, Cu-cumulus, Ns-nimbostratus, As-altostratus, Ac-altocumulus, De-vertically extends cloud, Cs-cirrostratus, Ci-cirrus, Clear-clear sky, MLwater-multilayer water cloud, MLice-nabivnoy ice cloud；The knowledge of this area image Other result is just without Filling power pixel.
Detailed description of the invention
Below in conjunction with the accompanying drawings and detailed description of the invention further illustrates the present invention.
The present invention proposes in a kind of remote sensing images multi layer cloud and individual layer cloud-type compressive classification and knows method for distinguishing, and it is for can comprehensively know Other multi layer cloud phase and the method for individual layer cloud cloud type, concrete implementation flow process is as shown in Figure 4.
In figure, each parameter is as follows: τ is optical thickness；PtopFor cloud top pressure；PsFor earth's surface pressure；PtropFor troposphere pressure；FMLIdentify for multi layer cloud；T8.5It it is the bright temperature of cloud layer under 8.5 mu m wavebands；T11It is that under 11 mu m wavebands, cloud layer is bright Temperature；T12It it is the bright temperature of cloud layer under 12 mu m wavebands；D8.5-11It is the bright temperature difference of cloud layer under 8.5 mu m wavebands and 11 mu m wavebands, D8.5-11=T8.5-T11；D11-12It is the bright temperature difference of cloud layer, D under 11 mu m wavebands and 12 mu m wavebands11-12=T11-T12.Additionally, in figure Common individual layer cloud refer to not for the vertical individual layer cloud extending cloud；Figure omits Filling power pixel (cannot sentence knowledge pixel)；ISCCP Threshold value means and utilizes cloud top pressure and optical thickness to judge the threshold value corresponding to cloud-type.
Concrete identification step is as follows:
Step-1: have cloud pixel to differentiate
According to cloud mask data, satellite image pixel is divided into cloud pixel, clear sky pixel and Filling power pixel and (knowledge cannot have been sentenced Pixel), and for there being cloud pixel to complete following cloud layer recognition step.
Step-2: individual layer and multi layer cloud rough sort and mark
Utilize earth's surface pressure Ps, troposphere top pressure Ptrop, cloud top pressure Ptop, opticalthicknessτ and multi layer cloud mark FMLAccording to Following decision condition carries out rough sort to cloud layer:
(1) Δ P is mettop> 0.6 and τ > 12 item judge that this pixel is as vertically extending cloud；
(2) Δ P is mettop>0.6, τ<12 and FML>=2 judge that this pixel is as multi layer cloud pixel；
(3) the pixel then preliminary judgement being all unsatisfactory for (1) (2) is common individual layer cloud pixel.
Step-3: individual layer cloud-type sophisticated category
By the pixel that preliminary judgement in step-2 is common individual layer cloud, utilize cloud top pressure Ptop, opticalthicknessτ parameter is according to ISCCP Individual layer cloud pixel is further subdivided into and vertically extends cloud, cirrostratus, cirrus, altostratus, altocumulus, nimbostratus, long-pending by threshold value Cloud, stratocumulus and the pixel of nine kinds of cloud type of stratus.
Step-4: the classification of cloud layer phase and mark
Step-2 will be judged to the pixel of multi layer cloud, utilize 8.5 mu m wavebands and bright temperature difference D of 11 mu m wavebands8.5-11With 11 mu m wavebands With 12 bright temperature difference D of mu m waveband11-12, according to following condition, multi layer cloud pixel is judged:
If meeting D8.5-11≥D11-12Then judge that pixel is as nabivnoy ice cloud pixel；
If meeting D8.5-11<D11-12Then it is judged to multilayer water cloud pixel.
Thus complete that remote sensing images have the discriminating of cloud pixel and the compressive classification of cloud layer and identification.
Embodiment:
Integrated multilayer cloud proposed by the invention illustrated below and the concrete application process of individual layer cloud-type recognition methods.
Use the observation time that on June 1st, 2015 (Julian calendar 152 days) Terra satellite, MODIS sensor is observed Track (the five minutes bands) data being spaced for the five-minute period of 02:30-02:35, have 5km resolution ratio 406 × 207 and 1km 2030 × 1,354 two kinds of specifications of resolution ratio.Fig. 5 is MODIS RGB pseudo color image, has the cloud of large area in this image Layer covers.It addition, the cloud type occurred in this image is the most, and there is the situation of multi layer cloud ply.Now utilize the present invention The method proposed carries out multi layer cloud phase to the 5km resolution satellite image data in region corresponding to Fig. 5 and individual layer cloud-type is known Not.
The cloud mask data [9] of MODIS image stores with decimal integer form, needs decimal integer is converted into two System by " bit ", figure place and numerical value according to " bit " corresponding to each mark judge, if the cloud corresponding to a pixel Binary number 2-0 " bit " (being followed successively by 76543210 to low bit number from a high position) value of the first character joint of mask data is 111, Then judge that this pixel is as clear sky pixel.Result is differentiated as shown in Figure 6 with the cloud mask corresponding to image in Fig. 5.Clear sky, possibility Clear sky and uncertain mark represent that the confidence level that pixel is clear sky is respectively 99%, 95% and 66%.To be labeled as may clear sky, Uncertain and have the pixel of cloud to be all classified as cloud pixel, continue follow-up classification and identification.
Fig. 7 shows the result of method proposed by the invention, MODIS multi layer cloud during wherein Fig. 7 (a) is step-2 Mark result figure, the multi layer cloud mark MLF value of MODIS is the integer in the range of 0-9, and wherein, 0 represents clear sky；1 generation List stratus；2-9 all represents multi layer cloud, and numerical value to be expressed as the most greatly the confidence level of multi layer cloud the highest, MLF is more than herein Pixel in 2 is all judged to multi layer cloud pixel；Fig. 7 (b) is the preliminary judgement result of individual layer cloud and multi layer cloud in step-2； Fig. 7 (c) is the ISCCP threshold determination cloud-type result in step-3 to individual layer cloud pixel；Fig. 7 (d) is that the present invention is final Multi layer cloud phase and the recognition result of individual layer cloud-type.
The result of multi layer cloud that Fig. 7 is proposed by the invention and individual layer cloud-type compressive classification and recognition methods, the symbol corresponding to colour code The explanation of number title is: St-stratus, Sc-stratocumulus, Cu-cumulus, Ns-nimbostratus, As-altostratus, Ac-altocumulus, and De-hangs down DS fuzz, Cs-cirrostratus, Ci-cirrus, Clear-clear sky, MLwater-multilayer water cloud, MLice-nabivnoy ice cloud；This district The recognition result of area image is just without Filling power pixel
Other replacement scheme of the present invention can complete goal of the invention equally:
(1) this specification is as a example by Terra satellite platform MODIS sensor remote sensing images, illustrates the basic skill of the present invention Art thinking, processing method and flow process, use the bright temperature characteristics of cloud layer under other observation platforms or its all band or reflection characteristic to replace In step-4, under 8.5 μm, 11 μm and 12 mu m wavebands, bright temperature characteristic also completes to identify cloud layer phase, but processing method and flow process without Substantial change need to be made；
(2) institute's bill of lading layer and the identification of multi layer cloud in this invention, it is possible to use other multi layer cloud pixel recognition methods Become, for example with the method discussed in PH algorithm (seeing document [7]) alternative steps-2, individual layer and multilayer can be completed equally The identification of cloud.

Claims (1)

1. in remote sensing images multi layer cloud and individual layer cloud-type compressive classification with know method for distinguishing, it is characterised in that the method Comprise the steps:
Step-1: have cloud pixel to differentiate
According to cloud mask data, satellite image pixel is divided into cloud pixel, clear sky pixel and Filling power pixel, this Filling power Pixel is for cannot sentence knowledge pixel, and for there being cloud pixel to complete following cloud layer recognition step；
Step-2: individual layer and multi layer cloud rough sort and mark
Utilize earth's surface pressure Ps, troposphere top pressure Ptrop, cloud top pressure Ptop, opticalthicknessτ and multi layer cloud mark FMLAccording to Following decision condition carries out rough sort to cloud layer:
(1) Δ P is mettop> 0.6 and τ > 12 item judge that this pixel is as vertically extending cloud；
(2) Δ P is mettop>0.6, τ<12 and FML>=2 judge that this pixel is as multi layer cloud pixel；
(3) the pixel then preliminary judgement being all unsatisfactory for (1) (2) is common individual layer cloud pixel；
Step-3: individual layer cloud-type sophisticated category
By the pixel that preliminary judgement in step-2 is common individual layer cloud, utilize cloud top pressure Ptop, opticalthicknessτ parameter is according to ISCCP Individual layer cloud pixel is further subdivided into and vertically extends cloud, cirrostratus, cirrus, altostratus, altocumulus, nimbostratus, long-pending by threshold value Cloud, stratocumulus and the pixel of nine kinds of cloud type of stratus；
Step-4: the classification of cloud layer phase and mark
Step-2 will be judged to the pixel of multi layer cloud, utilize 8.5 mu m wavebands and bright temperature difference D of 11 mu m wavebands8.5-11With 11 mu m wavebands With 12 bright temperature difference D of mu m waveband11-12, according to following condition, multi layer cloud pixel is judged:
If meeting D8.5-11≥D11-12Then judge that pixel is as nabivnoy ice cloud pixel；
If meeting D8.5-11<D11-12Then it is judged to multilayer water cloud pixel；
Thus complete that remote sensing images have the discriminating of cloud pixel and the compressive classification of cloud layer and identification.
CN201610237086.9A 2016-04-15 2016-04-15 Multi layer cloud and single layer cloud-type compressive classification and knowledge method for distinguishing in remote sensing images Active CN105913033B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610237086.9A CN105913033B (en) 2016-04-15 2016-04-15 Multi layer cloud and single layer cloud-type compressive classification and knowledge method for distinguishing in remote sensing images

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610237086.9A CN105913033B (en) 2016-04-15 2016-04-15 Multi layer cloud and single layer cloud-type compressive classification and knowledge method for distinguishing in remote sensing images

Publications (2)

Publication Number Publication Date
CN105913033A true CN105913033A (en) 2016-08-31
CN105913033B CN105913033B (en) 2019-05-14

Family

ID=56746227

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610237086.9A Active CN105913033B (en) 2016-04-15 2016-04-15 Multi layer cloud and single layer cloud-type compressive classification and knowledge method for distinguishing in remote sensing images

Country Status (1)

Country Link
CN (1) CN105913033B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109446899A (en) * 2018-09-20 2019-03-08 西安空间无线电技术研究所 A kind of cloud object detection method based on four spectral coverage remote sensing images
CN109541723A (en) * 2018-11-23 2019-03-29 中国气象科学研究院 A kind of analysis method of sleet temperature and humidity structure
CN109946235A (en) * 2019-02-26 2019-06-28 南京信息工程大学 The multi layer cloud inversion method of wind and cloud 4A meteorological satellite Multichannel Scan Imagery Radiometer
CN110261341A (en) * 2019-06-20 2019-09-20 中国矿业大学(北京) A kind of volcanic ash cloud detection method and system based on stationary weather satellite data
CN111340093A (en) * 2020-02-21 2020-06-26 上海眼控科技股份有限公司 Cloud identification method and device
CN114778454A (en) * 2022-03-30 2022-07-22 中国人民解放军国防科技大学 Multilayer cloud top height inversion method based on satellite-borne multi-channel radiation imager

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7058511B2 (en) * 2003-11-24 2006-06-06 The Boeing Company Sub-visible cloud cover assessment: VNIR-SWIR
CN102799903A (en) * 2012-07-18 2012-11-28 中国科学院自动化研究所 High-spectrum automatic cloud detection method based on space and spectral information
CN103093243A (en) * 2013-01-24 2013-05-08 哈尔滨工业大学 High resolution panchromatic remote sensing image cloud discriminating method
CN103150567A (en) * 2013-02-04 2013-06-12 北京理工大学 Optical remote sensing image cloud distinguishing method of integrating mode recognition and context knowledge
CN103926634A (en) * 2014-03-12 2014-07-16 长江水利委员会长江科学院 Daytime land radiation fog remote sensing monitoring method based on object-oriented classification

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7058511B2 (en) * 2003-11-24 2006-06-06 The Boeing Company Sub-visible cloud cover assessment: VNIR-SWIR
CN102799903A (en) * 2012-07-18 2012-11-28 中国科学院自动化研究所 High-spectrum automatic cloud detection method based on space and spectral information
CN103093243A (en) * 2013-01-24 2013-05-08 哈尔滨工业大学 High resolution panchromatic remote sensing image cloud discriminating method
CN103150567A (en) * 2013-02-04 2013-06-12 北京理工大学 Optical remote sensing image cloud distinguishing method of integrating mode recognition and context knowledge
CN103926634A (en) * 2014-03-12 2014-07-16 长江水利委员会长江科学院 Daytime land radiation fog remote sensing monitoring method based on object-oriented classification

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
J. JOINER ET AL.: ""Detection of multi-layer and vertically-extended clouds using A-train sensors"", 《ATMOSPHERIC MEASUREMENT TECHNIQUES》 *
叶晶 等: ""基于MODIS数据的白天多层云检测算法"", 《遥感技术与应用》 *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109446899A (en) * 2018-09-20 2019-03-08 西安空间无线电技术研究所 A kind of cloud object detection method based on four spectral coverage remote sensing images
CN109541723A (en) * 2018-11-23 2019-03-29 中国气象科学研究院 A kind of analysis method of sleet temperature and humidity structure
CN109946235A (en) * 2019-02-26 2019-06-28 南京信息工程大学 The multi layer cloud inversion method of wind and cloud 4A meteorological satellite Multichannel Scan Imagery Radiometer
CN109946235B (en) * 2019-02-26 2021-11-02 南京信息工程大学 Multilayer cloud inversion method of wind cloud 4A meteorological satellite multichannel scanning imaging radiometer
CN110261341A (en) * 2019-06-20 2019-09-20 中国矿业大学(北京) A kind of volcanic ash cloud detection method and system based on stationary weather satellite data
CN110261341B (en) * 2019-06-20 2020-10-02 中国矿业大学(北京) Volcanic ash cloud detection method and system based on stationary meteorological satellite data
CN111340093A (en) * 2020-02-21 2020-06-26 上海眼控科技股份有限公司 Cloud identification method and device
CN114778454A (en) * 2022-03-30 2022-07-22 中国人民解放军国防科技大学 Multilayer cloud top height inversion method based on satellite-borne multi-channel radiation imager

Also Published As

Publication number Publication date
CN105913033B (en) 2019-05-14

Similar Documents

Publication Publication Date Title
CN105913033A (en) Multi-layer cloud and single-layer cloud type integrated classification and identification method in remote sensing image
Ferreira et al. Tree species classification in tropical forests using visible to shortwave infrared WorldView-3 images and texture analysis
Pu et al. Mapping urban tree species by integrating multi-seasonal high resolution pléiades satellite imagery with airborne LiDAR data
Chen et al. Multi-source remotely sensed data fusion for improving land cover classification
Pelletier et al. Assessing the robustness of Random Forests to map land cover with high resolution satellite image time series over large areas
Summers et al. Visible near-infrared reflectance spectroscopy as a predictive indicator of soil properties
Shahi et al. A novel spectral index to automatically extract road networks from WorldView-2 satellite imagery
Ganivet et al. Towards rapid assessments of tree species diversity and structure in fragmented tropical forests: A review of perspectives offered by remotely-sensed and field-based data
CN102496034B (en) High-spatial resolution remote-sensing image bag-of-word classification method based on linear words
Oyama et al. Monitoring levels of cyanobacterial blooms using the visual cyanobacteria index (VCI) and floating algae index (FAI)
CN103034863A (en) Remote-sensing image road acquisition method combined with kernel Fisher and multi-scale extraction
CN107942411A (en) A kind of atmospheric visibility Forecasting Methodology
Hao et al. Potential of multitemporal Gaofen-1 panchromatic/multispectral images for crop classification: case study in Xinjiang Uygur Autonomous Region, China
Wei et al. Global satellite water classification data products over oceanic, coastal, and inland waters
Karimzadeh et al. A deep learning model for road damage detection after an earthquake based on synthetic aperture radar (SAR) and field datasets
Hidayati et al. A comparative study of various indices for extraction urban impervious surface of Landsat 8 OLI
Chen et al. A new method for automated monitoring of road pavement aging conditions based on recurrent neural network
Shahrisvand et al. A comparison of empirical and inteligent methods for dust detection using modis satellite data
Van der Werff et al. Combining spectral signals and spatial patterns using multiple Hough transforms: An application for detection of natural gas seepages
Karakizi et al. Spectral discrimination and reflectance properties of various vine varieties from satellite, UAV and proximate sensors
Chabrillat et al. Potential of hyperspectral imagery for the spatial assessment of soil erosion stages in agricultural semi-arid Spain at different scales
Zlobina et al. Algorithm for statistical analysis of multispectral survey data to identify the anthropogenic impact of the 19th century on the natural environment
Jin et al. Automated road pavement marking detection from high resolution aerial images based on multi-resolution image analysis and anisotropic Gaussian filtering
AU2021100533A4 (en) Soil salinity at Yellow River Delta Inversion Method based on Landsat 8
May et al. Spectral signature selection for mapping unvegetated soils

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant