IN84.

21RemoteSensingandGISforDisasterMitigationLaboratory

LabExcerciseReport#1 IntroductionofBasicToolandColorCompositionImages
By:WifandyRTPurbaSidadolog #114354 DisasterPreparedness,MitigationandManagement(DPMM) AsianInstituteofTechnologyAIT 2012

LaboratoryExercise1 INTRODUCTIONOFBASICTOOLANDCOLORCOMPOSITIONIMAGES Date :August28,2012 StudentsName:WifandyRaymondTobiasPurbaSidadolog ID :114354 LabInstructor :[Link] INTRODUCTION RemoteSensingisthescienceandartofobtaininginformationaboutanobject,area,orphenomenonthrough the analysis of data acquired by a device that is not in contact with the object, area, or phenomenon under investigation[1]. In relation with a disaster cccurence, the technology of Remote Sensing is widely used, if not said,averysignificantaspectduringtheprocessofpreparedness,mitigationandresponseofthedisaster [2].In relation withdisastermanagement,theearthobservationthrough satellite plays a [Link] datasensedthroughspacetechnology(satellite)hasbeenwidelyusedduringthedisastermanagementprocess fori.e.: Monitoringofnaturalhazardspotentialfornaturaldisasteroccurence([Link]) RapidAssessmentofdisasterimpact Locatingplacesforrefugees Decissionmakingforinfrastructurerepairpriorities Andsoon. However, to deliver its purpose, the geospatial data received from satellites need to be processed through certaincomputertechnologybyacompetentperson,beforeitcanbeusedeasilybydecisionmakers. ERDASIMAGINE9.3isisaremotesensingapplicationwithrastergraphicseditorabilitiesdesignedbyERDASfor geospatial applications. This software allows the user to prepare, display, and enhance digital images. By manipulating the raw data using this software, it is possible to analyze some objects which are normally not visible[3]. OBJECTIVES ThisLabSessionisintendedtomakestudentsfamiliarwiththisERDAS9.3ImageProcessingSoftwareandwith the multispectral data. The image processing is used to process the image for the purpose of Disaster Management (Preparedness, Mitigation, and Response). Student will open and manipulate individual band of LandsatTMdatausingERDASIMAGINE. DATATOBEUSED TheimageusedduringthislabisLANDSAT5TMImage.TheTMImagealwayscoversa185x185kmarea[4]. TheLANDSAT5TMImagewasstoredintheAITnetworkaddress://ibmserv/public_dellserv/ Landsat5isaSunSynchronousSatelitelaunchedonMarch1,1984equippedwithtwomultispectralsensors:a Multispectral Scanner (MSS 4 spectral bands) and a Thematic Mapper (TM 7 spectral bands). The characteristicofLandsat5is: OrbitType :SunsynchronousPolar. Altitude :705km OrbitPeriod :98.9min TotalFOV :14.920(185x185km)

Inclination :98.20 Repeatcycle :16days(233revolutions) Equatorialcrossingtime:10.00AM TheTMSensor7SpectralBandcharacteristics[2],[5]. Band Spectral Resolution Band(m) 1Blue 0.450.52 30x30m

Use Ground/plantsdifferentiation,coastalzones, Bathymetricmapping,distinguishingsoilfrom vegetationanddeciduousfromconiferous vegetation Vegetation: Emphasizes peak vegetation, which is usefulforassessingplantvigor Plantspeciesdifferentiation,discriminates vegetationslopes Emphasizesbiomasscontentandshorelines Snow/clouddifferentiation,Discriminatesmoisture content of soil and vegetation; penetrates thin clouds Thermalmappingandestimatedsoilmoisture Hydrothermally altered rocks associated with mineraldeposits

2Green 3Red 4NearIR 5ShortwaveIR

0.520.60 0.630.69 0.760.90 1.551.75

30x30m 30x30m 30x30m 30x30m

6ThermalIR 7ShortwaveIR

10.412.5 2.082.35

120x120m 30x30m

METHODOLOGY InthisLabsession,studentsareparticularlyassignedtoachievetwogoalsofskillset: 1. DisplaySatelliteImageinColorCompositeofTrueColor,FalseColor,PseudoColor,GrayscaleandRelief[5] Acolorcompositeisthecolorofimagegeneratedbycomposingthreeselectedmultibandimageswiththe useofthreeprimarycolor(RGB).[Link] differentcolors,dependingbytheselectionofthreebandimagesasignedtothethreeprimarycolors. ThetruecolorcompositeiswhenthreefiltersofR,GandBareassignedtothesamespectralregionsofred (Band3),green(Band2)andblue(Band1),[Link] [Link] (1,2,3)orotherinvisibleregions(suchasIRBands)areassigned/allocatedtoRGBFilters FordisplayingimageswithdifferentBandcombinationassignedtotheRGBfiltersweuseToolsfromView [Link] Openafileandchoosetheoriginalimage([Link]) UsetheRastertab,choosedisplayasTrueColor. [Link],RedisassignedwithBand3, GreenwithBand2,[Link],wearetryingseveralcombination toseetheresults Pseudocolorimageisdisplayedwhendifferenceofcolorsareassignedtothesubdividedgreyscaleofa singleimage. Fordisplayingapseudocolorimage,[Link]

Openafileandchoosetheoriginalimage([Link]) UsetheRastertab,choosedisplayasPseudocolorwithLayer6(ThermalIR) OntheViewerwindowchooseRaster,thenselectAttributes,Edit,thanColors ChangeColorstoseedifferentcombinations.

FordisplayingaGrayscaleimage,[Link] Openafileandchoosetheoriginalimage([Link]) UsetheRastertab,choosedisplayasGreyscalewithLayer4(NearIR) FordisplayingaReliefimage,[Link] Openafileandchoosetheoriginalimage([Link]) UsetheRastertab,choosedisplayasReliefwithLayer4(NearIR) 2. CreateSubsectionadataset[5] Subsettinganimageisaprocesstodelimitasmallareaofinterestfromtheinputimagethatcoverslarger areathannecessary. Forcreatingasubsetdata,[Link]: SelectSubsetImage [Link] [Link] follows: ULX :659995 ULY :1565845 LRX :693832 LRY :1542288 Clickok. RESULTANDDISCUSSION BelowarethescreenshotsoftheoutputimagesduringtheLaboratoryExcerciseonAugust28,[Link] severalcombinationexcercisedduringtheLabsession(morethan20combinations),whichsomeofthe screenshotsarerepresentedbelow(notallcombinationsscreenshotsaredisplayedinthisreport). a. OutputofTrueColorComposite(3,2,1BandsassignedtotheRGBFilters). ThisisthecombinationofBand3,2and1assigned to the RGB Filter. With this image we can easily identify the River (BlueGreenish), Buildings (White rectangulars), vegetations (Green Gradations), and Barelands (Brownish). The different greens (gradation) can be interpreted as different kind or differentconditionofthevegetations.

b. OutputofFalseColorComposite(severalcombinationsofBandassignedtotheRGBFilters)

432BandCombination

532BandCombination

721BandCombination

645BandCombination

 253BandCombinations

754BandCombinations

ByselectingdifferentofBandCombination,wecanidentifythedifferentobjectswhichareindifferentcolor [Link],excepttheriverstreamwhichshapeisveryunique,itisdifficulttointerpretetheotherobjects andmeaningofthecolors. c. OutputofPseudoColorImage PseudoColorImageofLayer6(ThermalIR)inCyantoRedGradation

Theuseofpseudocolorisespeciallyusefulfordisplayingdatathatarenotreflectionmeasures,forexamplethe Thermal infrared red data. By displaying this Pseudocolor Image of Layer 6 (Thermal IR) we can see the the associationofcoldwarmwithredyellowdata,whichismoreintuitive[6].

d. OutputofGrayColorImageandReliefImage

GrayScaleImageOutput

ReliefImageOutput

Frombothimageoutput,wecanidentifytheboundaryandshapeoftheobjects,however,excepttheriver streamwhichshapeisveryunique,itisdifficulttointerpreteandconcludetheactualformoftheobject, whetheritisabuilding,afarm,afields,orbarelands. e. OutputofSubsettingtheimage Subsetimage([Link])comparedwithsourceimage([Link])inTrueColorComposition. A B

Byinputtingthecoordinateoftheedges/cornerinthesubsetparameter,wecancreateasubsetimagecovering ourareaofinterestanddiscardotherpartoftheimage.

CONCLUSION WecandisplaythesatelliteimagefromLandsat5ThematicMapper([Link])inseveralways,usingthe RasterOptions(TrueColorImage,FalseColorImage,PseudoColorImage,GrayColorImage,andReliefImage). Thisoptionisveryimportantintheconceptofdatavisualizationandcontrastenhancement,asitgivesusthe ability to choose the best combination of TM Spectral Bands assigned to the RGB filter. By having the knowledge of the Landsat 5 TM Bands characteristic, we can analyze the output images and interprete the [Link],dependsonthepurposeofimageanalysis. For Land Surface Observation Analysis, the combination I like the most is the True Color Composition 321. UsingthisTrueColorComposition,wecaninterpretethebuildings,theriverstreamandthevegetationastheir real color. The different gradation GreenBlue of the water body and the different gradation of the Green vegetation represents some divergent condition of the water body and vegetations. Also the Brownish area [Link] area monitoring, rapid damage assesments, flood spread analysis, etc. This image output accesibility is also criticalduringthedecissionmakingprocessofPublicAuthoritiesrelatedwithdisaster. ByperformingImageSubsettingontheoriginalimage,wecanhaveasmaller/delimitedareaofinterest,without loosing its resolution and band properties. It is important to select the accurate area of interest, to be more focuswhenworkingwiththeimage. REFERENCES [1] Lillesand,T.M.,Kiefer,R.W.,andChipman,[Link],[Link] Wiley&Sons,NewYork,NY,2004 [2] Nirupama,Simonovic,S.P.,RoleofRemoteSensinginDisasterManagement,PaperSeriesNo.21,ICLR Research,2004. [3] [Link] [4] [Link] [5] IN84.21RemoteSensingandGISforDisasterMitigationLabManualprovidedbytheFaculty [6] Kerle,N.,Janssen,L.F.,Hurneeman,G.H.,PrinciplesofRemoteSensing,[Link],Enshede,The Netherlands,2004