791 PHOTOGRAMMETRIC45075820912867111141183222254291333374402433465478515 ENGINEERING97975820912769109122164191218250263305345 &1338758209136 REMOTE138875820913259101145174201 SENSING16037582091265395121134176215 March502375930785592121153191 2014523875930784181120155 239562575910803570105 Ensemble134658603591382713865157158479111041 Learning24475860359122252383467600664797931 with34395860359175239324456 Multiple39555860359209340404490554685749878 Classifiers95489503591522163474625776417097779069901107 and21228950359131264395 Polarimetric257789503591472743384695536178179471032111611801304 Features39428950359134258390475606690820934 for4937895035977205289 Polarized123212050359146273337469553617721850981 SAR227312050359141286446 Image27791205035970270401535665 Classification35041205035915221734846257764170977790110331118118213111444 Alim.1343180101145579102177200 Samat,15651801011452101175224256279 Peijun18671801011454101125149198247 Du,21371801011458107130 Muhammad22891801011478127175224298373421470 Hasan27811801011459108150199248 Ali3052180101145578102 Baig,31771801011458107131181204 Sumit34031801011453101176199231 Chakravarty,36571801011456105154203234283322370401434474495 and4175180101144898146 Liang4344180101144569118167217 Cheng45831801011457106154203253 Abstract450212101515197137167196241284314 Polarimetric4502252010755105133183218247327372402437465510 SAR985226808544100148 (PolSAR)1158225201073011882268086448410715120525314422252010730 image14972252010728108158203248 processing1770225201075590140185230270310339394439 has22342252010755105145 become2404225201075095140190270315 a27442252010750 hot28192252010755105135 research450235201073580120165215250295350 topic825235201073080135163208 in1058235201072984 SAR1167236808544100148 remote1340235201073580160210240285 sensing1650235201074085140180208263308 field1983235201072855100129184 in2192235201072883 recent2300235201073580125170225255 years.2580235201074792142177217242 However,4502452010770120188233280325360380 due8552452010755110155 to1035245201073080 the1140245201073085130 complexity1295245201074595175230259304354382412459 of1779245201075080 the1884245201073085130 image20392452010728108158203248 and23122452010750105160 limited2497245201072857137165195240295 availability450255201075097147175203254304332360389419465 of940255201075080 advanced10452552010750105152202257302347402 techniques,1472255201073075120175230258308363408448473 PolSAR197025680864484107151205253 image22492552010728108158203248 processing2522255201075590140185230270310450265201072883128 is603265201072969 still69726520107407098126155 a8772652010750 challenging9522652010745100150178206251306351380435480 issue.1457265201072868108163208233 In1715265201073085 this1825265201073085113153 paper20032652010755105160205240 the2268265201073085130 suggestion2423265201074095140185230270300329379434 of450275201075080 ensemble5552752010745100140185265315343388 learning968275201072974124159214242297342 (EL)1335275201073013652768086448414492752010730 is1504275201072868 introduced1597275201072984114149199254309354399454 into2076275201072883113163 PolSAR226427680864484107151205253 image25432752010728108158203248 classification450285201074573123163203232259287332382412440490545 by1020285201075097 integrating1142285201072883113158203238288318347402446 various1614285201074696131160210265305 polarimetric19442852010755105133183218246326371401436465510 features2479285201073075125155210245290330 and4502952010750105160 multiple6352952010780135163193222277305350 classifiers.1010295201074573123163203232259287332367407432 The14672952010760115160 most16522952010780130170200 popular18772952010755105160215243293328 ensemble22302952010745100140185265315344389 learning2644295201072873123158213241296341 methods,4503052010780125155210260315355380 including855305201072883128157212267295350395 Bagging,12753052010758108153198227282327352 AdaBoost,16523052010768123173231281332371401426 and21033052010751106160 Rotation22893052010760110140190220248298353 Forest26673052010753103138183223253 are450315201075085130 adopted6053152010750105155210240285340 to970315201073080 combine1075315201074595175225254309354 multiple14543152010780135163193221276305350 classifiers1829315201074573123163203231259286331366406 and22603152010750105160 polarimetric24453152010755105134184219247327372402437465510 features.450325201073075125155210245290330355 The8303252010760115160 proposed1015325201075590140195245285330385 classification1425325201074573123163203232259287332382412440490545 scheme1995325201074085140185265310 is2330325201072868 evaluated2423325201074592142171226276306351406 on28543252010750105 three450335201073085120165210 real685335201073580130158 PolSAR86833680864484107151205253 data.11473352010755105135185210 Experimental13823352010755105160205240268348393448478528557 results1964335201073580120175203233273 shows2262335201074095145213253 that2540335201073085135165 the2730335201073085130 covariance450345201074595142192227255305360405450 and9253452010750105160 coherence1110345201074595150195230275330375420 features1555345201073075125155210245290330 can1910345201074595150 give2085345201074574120165 better2275345201075095125155200235 performance253534520107551001351652152504503552010780130185230275 than750355201073085135190 other965355201075080135180215 polarimetric12053552010755105133184219247327372402437465510 decomposition17403552010755100145195275330380420449479507557612 features,2377355201073075125155210245290330355 and27573552010750105160 complementary450365201074595175230258303383428483513563598645 between1120365201075095125193238283338 different1483365201075584114144189224269324354 polarimetric18623652010755105133183218247327372402437465510 decomposition2397365201075510014519527533038045037520107406898127177232 features707375201073075125155210245290330 improving10623752010728108163198248295323378423 the1510375201073085130 classification1665375201074573123163203232259287332382412440490545 performance.22353752010755100135165215250330380435480525550 Although450385201076897127182232287332387 a8623852010750 weak9373852010768113163213 classifier1175385201074573124163203232259287332367 gives1567385201074573120165205 unsatisfactory17973852010755110150200230258298328378423453503538585 classification2407385201074573123163203232259287332382412440490545 accuracy450395201075095140195230280325372 on8473952010750105 polarimetric9773952010755105133183218247327372402437465510 decomposition15123952010755100145195275330380420448478507557612 features,2149395201073075125155210245290330355 the2529395201073085130 performance26843952010755100135450405201073080115195245300345390 can865405201074595150 be1040405201075095 highly1160405201075584129184212259 improved14444052010728108163198248295340395 by1864405201075096 using1985405201075595124179224 EL223440680864484 strategies.2343405201074070105155185230275303348388413 Introduction450431101512470100129174219265308337360405451 Land4504442010755100155209 cover684444201074595145190224 classification933444201074575120160199229256284329374403433483538 is1496444201072969 one15904442010750105150 of1764444201075080 the1869444201073084129 primary2023444201075590120199244279329 objectives2377444201075099124169214244274323368408 of2810444201075079 remotely450454201073580160210239284314364 sensed839454201073984139179224278 image11424542010730110155200244 processing,1411454201075590140185229269309339393438463 analysis,18994542010745100144174224264294333358 and22824542010745100155 practical2461454201075590135180209239284329359 application.2845454201074499450464201075585115159204234264314368393 Synthetic8684642010755105160189244289319349393 Aperture12864642010770125170205234289324369 Radar16794642010761105160205240 (SAR)19444642010729197346580854410014821214642010730 opens21764642010750104149204244 new2445464201075599174 opportunities26444642010750105159209244274329450474201075585115144189229 for704474201073079114 land843474201073075130184 cover1052474201074595145190224 classification1301474201074575120159199229256284329373403433483538 owing18634742010750125155210255 to2142474201073080 its224747420107306099 day23714742010755100150 and25454742010745100155 night,2725474201075584129184214239 all-weather,450484201074575105134209254299329383428463483 and958484201074599154 high-resolution1137484201075585129184214249294334383413468498528577632 observation17944842010750100140184219269314344373423478 capability2297484201074590144189239269299328358408 to2730484201073080 the2834484201073085130 Earth’s4504942010755100135164219244279 surface.754494201074094129159204249293318 But10974942010760115145 due12664942010755110155 to1446494201072979 the1550494201073085130 low1704494201073080155 information1884494201072984114164199279323353383433488 content2397494201074494149179224278308 of2730494201075080 an2835494201074499 individual450504201073085140169219249304359403433 SAR908505808544100148 image,10815042010730110154199244269 single-polarization1375504201074069124169199244273328378408453487517562607637667716771 or2171504201075085 single-band2281504201073969124169199243273323368423477 SAR2783505808544100148 data31582160010755100130174 cannot3356216001074590145200249279 provide3660216001075590140189219274319 highly4004216001075585130185215265 accurate4292216001074590135190224269299344 land4661216001073074129184 cover4870216001074594144189224 classification5120216001074574119159199229256283328373403433482537 maps31582260010780125180219 (Herold34012260010730100145180229259314 et3740226001074575 al.,38392260010750793918226001072550 2004;3993226001075099149199224 Lee4242226001075599144 et4411226001074575 al.,45102260010750794588226001072550 2002).4663226001075099149199229254 On49412260010770125 the5091226001073085130 contrary,5245226001074595150180214259294344362 PolSAR315823770854484108152207255 images34382360010730110154199244284 contain3747236001074594149179224254308 a40802360010745 large4150236001073075109154199 amount43742360010745124174229284314 of4712236001075080 potential48172360010755105135179234264294339368 information5210236001073085115164199279324354384433488 which31582460010775130160204259 is3442246001073070 directly3536246001075585120165210239269319 related3880246001073579109154184229283 to4188246001073080 physical42932460010755109159199229274319348 properties4666246001075590140194239274304334378418 of5109246001075080 natural52132460010755100130185220264294 media553224600107801253158256001075585130 and33122560010745100155 backscattering3492256001075094139189229274318348378423458488542587 mechanism.41042560010780125169224269324354393473498 Therefore,46272560010760115159194239269319354398423 full5075256001073085114144 polarized52442560010755105135179214244289334388 SAR3158267708544100148 imagery33302660010730110154199244279329 classification3684266001074474119159199228256283328373403432482537 has42462660010755100139 become4410266001075095140190269314 a47492660010745 popular48192660010755104159214244289323 and51672660010745100155 challenging53472660010744991441742042483033483157276001073085130 problem3311276001075590140190220264344 in3680276001073085 recent3790276001073479124169224253 years4068276001075095140175214 due43072760010755110155 to4486276001073080 its459127600107306099 potential47152760010755105135179234264294339368 applicability51082760010745100155185214259304354384413443473523 in3157286001073085 different3266286001075585115145190224269324354 fields.364528600107275599129184224249 With39182860010790117146201 the4144286001073085130 goal4298286001074595140170 of4492286001075080 achieving4597286001074590144174219269299354398 high5020286001075585130184 classification52292860010745751201591992292562843293733157296001073060110164 accuracy,3346296001074590135190224269314364382 many37532960010780125179229 supervised,4007296001074095149194229279309349393448473 unsupervised,45052960010755109149204259304339388418458503558582 and51122960010745100155 semi-supervised529229600107398416419422426331831573060010755100135184214254299354 classification3536306001074474119159199228256283328373403432482537 methods40983060010780125154209259314354 have44763060010755100150195 been4695306001075095140195 developed.4915306001075599149194224274328373428453 Usually,53933060010770109164209239269318337 the3157316001073085130 adopted33113160010745100150205235279334 features3670316001073075119149204239284323 in4018316001073085 classification4128316001074574119159199229256283328373403433482537 include4690316001073085129159214269314 scattering5029316001073984129159189233268298353398 matrix54513160010780125155190220269 elements,3157326001074575120199244299329369393 covariance3575326001074595145190225254299354399444 matrix40433260010780125155190219269 elements4337326001074575120199244299329369 and47303260010745100155 coherent4910326001074594149194229274329358 matrix52933260010780125155189219269 elements,3157336001074575120199244299329369393 and35753360010745100155 polarimetric37543360010755105135180215244324369399434464508 decomposition42873360010755100145195274329379419449478508558613 features4925336001072974119149204239283323 (Cloude5273336001073090119169224279324 et3157346001074575 al.,32563460010750793335346001072550 1997;3410346001075099149199224 Du36593460010765119 et3803346001074575 al.,39023460010750783981346001072550 1996;4056346001075099149199224 Krogager,4305346001076599149194239284329363383 m47133460010780 1994;48183460010750100149199224 Qi50673460010770100 et5191346001074575 al.,52903460010750785368346001072550 2012;5443346001075099149199224 Song31573560010755105160204 et3387356001074575 al.,34863560010750783564356001072550 2007;3639356001075099149199224 Wang38883560010790130184229 et4142356001074575 al.,42413560010750784319356001072550 2013).4394356001075099149199229254 Generally,32773660010765110165210245290320350400418 developing37203660010755100150195225275330360415460 advanced42053660010745100150195250295340395 classifiers,4625366001074575120160200230258285330365405430 or5080366001075085 optimizing519036600107501051351652452753203157376001073085130 the3312376001073085130 input3467376001073085140195225 features3717376001073075120150205240285325 or4067376001075085 adopting41773760010745100150205235265320365 novel45673760010755105155200230 pattern48223760010755100130160205240295 recognition5142376001073580125175220275305335365415470 methods31573860010780125155210260315355 are3537386001074580125 the3687386001073085130 three3842386001073085120165210 common4077386001074595175255305360 ways44623860010775120170210 get4697386001074590120 better4842386001075095125155200235 classification5102386001074575120160200230257285330375405435485540 performance.31573960010755100135165215250330375430475520545 It3727396001073060 is3812396001073070 worth39073960010775125160190245 noting41773960010755105135165220265 that,4467396001073085130160185 finding4677396001072755110165195250295 the4997396001073085130 optimal51523960010750105135165245290320 features54973960010730751203157406001073085120165205 and33874060010745100155 best3567406001075095135165 classifier3757406001074575120160200230257285330365 is4147406001073070 always4242406001074575150195245285 hard,45524060010755100135190215 maybe47924060010780125175225270 even5087406001074595140195 impossible.530740600107301101652152552953253157416001075080125150 However,33324160010770120195240290335370390 a37474160010745 better3817416001075095125155200235 classification4077416001074575120160200230258285330375405435485540 result4642416001073580120175205235 may49024160010780125175 be5102416001075095 achieved5222416001074590145175220270315370 by56174160010750100 combining3157426001074595175225255310340395440 multiple36224260010780135165195225280310355 classifiers4002426001074575120160200230257285330365405 and44324260010745100155 various4612426001075095130160210265305 features4942426001073075120150205240285325 by52924260010750100 using5417426001075595125180225 ensemble31574360010745100140185265315345390 learning3572436001073075120155210240295340 (EL)3937436001073039674377085448840554360010730 strategies.4110436001074070105150180225270300345385410 Since3278446001075585140185230 “strong3533446001074585115150200255300 and38584460010745100155 weak40384460010775120165215 learn4278446001073075120155210 ability4513446001074595125155185215265 are4803446001074580125 equivalent”4953446001074595150180230275305350405435480 was54584460010775120160 proofed3158456001075590140190220265320 by35034560010750100 Robert36284560010760110160205240270 E.3923456001075580 Schapire40284560010755100155200255285320365 (Schapire,4418456001073085130185230285315350395420 1990),48634560010750100150200230255 more51434560010780130165210 options53784560010751106135165215270310 are3158466001074580125 available3308466001074595140170200245295325370 to3703466001073080 get3808466001074590120 an39534660010745100 improved40784660010730110165200250300345400 classification4503466001074575120160200230257285330375405435485540 performance50684660010755100135165215250330375430475520 through3158476001073085120170225270325 EL,35084777085448835954760010725 which36454760010775130160205260 can3930476001074590145 integrate4100476001073085115160205240285315360 the4485476001073085130 benefits4640476001075095150195222250280320 of4985476001075080 different5090476001075585115145190225270325355 features,54704760010730751203157486001073085120165205230 classifiers,3412486001074575120160200230257285330365405430 and38674860010745100155 combination4047486001074595175225255310355385415465520 methods.45924860010780125155210260315355380 As49974860010770110 an51324860010745100 advanced52574860010745100150195250295340395 machine31574960010780125170225255310355 learning3537496001073075120155210240295340 framework,3902496001073065110190235310360395445470 EL439849770854488 is4510496001073070 able4605496001074595125170 to4800496001073080 boost49054960010750100150190220 weak51504960010775120165215 learners,53904960010730751201552103158506001074580120145 working33285060010775125160210240295340 slightly3693506001074070100145200230260310 or4028506001075085 obviously41385060010750100150180230285325355405 better4568506001075095125155200235 than4828506001073085130185 random5038506001073580135190240320 guessing538350600107451001451852253158516001073085130 to3313516001073080 strong3418516001074070105155210255 learners,3698516001073075120155210255290330355 thus4078516001073085140180 making42835160010780125175205260305 accurate4613516001074590135190225270300345 predictions.4983516001075590135190220265295325375430470495 In3278526001073085 spite3388526001074095125155200 of3613526001075080 the3718526001073085130 considerable3873526001074595150190220275320355400450480525 amount44235260010745125175230285315 of4763526001075080 work48685260010775125160210 on51035260010750105 the5233526001073085130 use5388526001075595140 of5553526001075080 EL315853770854488 for3270536001073080115 remote3410536001073580160210240285 sensing3720536001074085140180210265310 image40555360010730110155200245 classification4325536001074575120160200230257285330375405435485540 in4890536001073085 the5000536001073085130 past51555360010755100140170 years,5350536001075095140175215240 few3157546001073075150 applications33325460010745100155185215260305335365415470510 of3867546001075080 EL397254770854488 to4085546001073080 PolSAR419054770854484108152208256 image44715460010730110155200245 classification4741546001074575120160200230257285330375405435485540 could5306546001074595150180235 be5566546001075095 found3158556001073080135190245 in3428556001073085 literature353855600107306090135170215245300335380 (Jiong394355600107306090140195240 et4207556001074575 al.,43075560010750784386556001072550 2008;44615560010750100150200225 Loosvelt47115560010755105155195245290320350 et5086556001074575 al.,518655600107507810352895560010725 2012;53395560010750100150200225 Min31585660010785115170 et3352566001074575 al.,34525660010750783531566001072550 2009;36065660010750100150200225 She38565660010755110155 et4036566001074575 al.,2007).41365660010750784214566001072550100150200250280305 In4544566001073085 this4654566001073085115155 paper,48345660010755100155200235255 EL511556770854488 techniques52285660010730751201753158576001075585135190235275 such3458576001074095140195 as3678576001074585 Bagging,37885760010760105150195225280325350 AdaBoost,41635760010770125170230280330370400425 and46135760010745100155 Rotation47935760010760110140185215245295350 Forest51685760010755105140185225255 are5448576001074580125 introduced3158586001073085115150200255310355400455 to3638586001073080 PolSAR374258770854484108152208256 image40245860010730110155200245 classification4294586001074575120160200230257285330375405435485540 based4859586001075095135180235 on51195860010750105 different5249586001075585115145190225270325355 polarimetric31575960010755105135180215245325370400435465510 features.3692596001073075120150205240285325350 To40675960010760102 evaluate4194596001074595140170225270300345 the4564596001073085130 performance47195960010755100135165215250330375430475520 of5264596001075080 various536959600107509513016031576060010750105145 EL332860770854488 methods,34416060010780125155210260315355380 some3846606001074090170215 benchmark4086606001075095150195250330375410460 classifiers4571606001074575120160200230257285330365405 such5001606001074095140195 as5221606001074585 support5331606001074095150205255290320 vector3157616001075095140170220255 machine34376160010780125170225255310355 (SVM),381761600107303848617708544961644011616001073055 logistics4091616001073080125155195225255300340 decision44566160010755100145175215245295350 tree,4831616001073065110155180 simple5036616001074070150205235280 logistic5341616001073080125155195225255300 decision31576260010755100145175215245295350 tree,3532626001073065110155180 and37376260010745100155 k-Nearest3917626001075080150195240275320360390 Neighbor43326260010770115145190245295345380 are4737626001074580125 experimented4887626001074595150195230260340385440470515570 and54826260010745100155 compared.3157636001074595175230275310355410435 Besides,36176360010760105145175230275315340 different3982636001075585115145190225270325355 ensemble43626360010745100140185265315345390 strategies4777636001074070105150180225270300345385 for5187636001073080115 classifier5327636001074575120160200230257285330365 combination3157646001074595175225255310355385415465520 such3702646001074095140195 as3922646001074585 voting,40326460010750100130160215260285 stacking4342646001074070115160210240295340 generation,4707646001074590145190225270300330380435460 and51926460010745100155 multiAlim.Samat53726460010780135165195225450528701077099128208232286330409454483 and957528701074498152 Peijun1133528701075599128152207261 are1418528701074478123 with15645287010775104133187 the1776528701072983127 Key19275287010765109158 Laboratory21085287010755100149198232276306355389438 for2570528701073079113 Satellite2706528701075510012917320245053870107305988132 Mapping6065387010784129183237266320365 Technology994538701076096140194248297326376420469 and1487538701074498153 Applications16635387010770124178207236280325354383432486525 of2211538701075180 State2314538701075584128157201 Administration25385387010770125204233287316356385419463450548701073059108163 of636548701075079 Surveying,7395487010755109143192236286315369413437 Mapping12005487010785129183237266321365 and1589548701074498153 Geoinformation17655487010765109158187242271320354433478507536585639 of2428548701075079 China,25305487010760115144198242266 Nanjing28205487010770114168450558701072554108153 University,6265587010770124153203247281320349378428445 Kunshan10955587010765119173212266310365 Building,14835587010760114143173227256310354378 Nanjing18855587010770114168192222276320 University22285587010770125154203247281321350379428 Xianlin2679558701076595139193222251305 Campus,4505687010760104183238292331355 No.163,8295687010770119143192241291315 Xianlin1167568701076594139193222251305 Dadao,14965687010765109163207256281 Qixia1800568701077099148178222 District,2045568701076594134163197226270299324 Nanjing23925687010770114168193222276320 City,2736568701076089118167185 Jiangsu450578701073059103158202241295 Province,768578701075590139188217271316360384 China11755787010760115144198242 2100231441578701075099148197247296 (alim.smt@gmail.com).1761578701072973102132211235274353383449493572617646675699743792872901925 Muhammad4505937010785140195240320400445500 Hasan9755937010770115155200255 Ali12555937010770100130 Baig14105937010760105135180 is1615593701073070 with17105937010775105135190 the1925593701073085130 Institute2080593701073085125155185215270300345 of2450593701075080 Remote25555937010760105185235265310 Sensing4506037010755100155195225280325 Applications,8006037010770125180210240285330360390440495535560 Chinese13856037010760115145200245285330 Academy17406037010770115160215260340390 of2155603701075080 Sciences,22606037010755100130175230275320360385 Hyperspectral26706037010770120175220255450613701074095140185215250295325 Remote8006137010760105185235265310 Sensing11356137010755100155195225280325 Laboratory14856137010755100150200235280310360395445 (HRS),19556137010730100160215245270 No.20,22506137010770120145195245270 Dutun25456137010765120150205260 Lu,28306137010755110135 Chaoyang4506237010760115160210260305360405 District,880623701076595135165200230275305330 Beijing,12356237010760105135160190245290315 China,15756237010760115145200245270 100101.18706237010750100150200250300325 Sumit4506387010755110190220250 Chakravarty7256387010760115160210245290340385420450500 is1250638701073070 with13456387010775105135190 the1560638701073085130 Department17156387010765110165210245275355400455485 of2225638701075080 Engineering23306387010755110155185240285330365395450495 and28506387010745100155 Computer4506487010760110190245300330375410 Science,8856487010755100130175230275320345 New12556487010770115190 York14706487010765108143193 Institute1688648701073085125155185215270300345 of2058648701075080 Technology,2163648701076097142197252302332382427477495 90326846487010750100150 Tree450658701076087132177 Creek652658701076095140185235 Parkway,9126587010755100135185260305355373 Lawrenceville,13106587010755100175210255310355400450480510540585610 GA,19456587010765135160 30043.21306587010750100150200250276 Liang450673701075585130185230 Cheng7056737010760115160215260 is990673701073070 with10856737010775105135190 the1300673701073085130 Key14556737010765110160 Laboratory16406737010755100150200235280310360395445 for2110673701073080115 Satellite22506737010755100130175205235265295340 Mapping26156737010785130185450683701075585140185 Technology660683701076097142197252302332382427477 and11626837010745100155 Applications13426837010770125180210240285330360390440495535 of1902683701075080 State2007683701075585130160205 Administration22376837010770125205235290320360390425470500530580635 of450693701075080 Surveying,5556937010755110145195240290320375420445 Mapping10256937010785130185240270325370 and14206937010745100155 Geoinformation16006937010765110160190245275325360440485515545595650 of2275693701075080 China,23806937010760115145200245270 Kunshan26756937010765120175450703701074095140195 Building,6707037010760115145175230260315360385 Nanjing10807037010770115170195225280325 University14307037010770125155205250285325355385435 Xianlin1890703701076595140195225255310 Campus,22257037010760105185240295335360 No.163,26107037010770120145195245295320 Xianlin450713701076595140195225255310 Dadao,7857137010765110165210260285 Qixia10957137010770100150180225 District,1345713701076595135165200230275305330 Nanjing17007137010770115170195225280325 City,2050713701076090120170188 Jiangsu2263713701073060105160205245300 Province,2588713701075590140190220275320365390 China4507237010760115145200245 210023.7207237010750100150200250300325 Photogrammetric37306657010755110160190240285320365445525570600635665710 Engineering44656657010755110155185240285330365395450495 &49856657010760 Remote50706657010760105185235265310 Sensing54056657010755100155195225280325 Vol.40466765010765108138163 80,42346765010750100125 No.43846765010771121146 3,4555676501075075 March46556765010785130165210265 2014,49456765010750100150200225 pp.51956765010755110135 239–251.53556765010750100150200250300350375 0099-1112/14/8003–239472968730107501001502002302803303804304655155656006507007518018519019511001 ©38277001010767 201439197001010750100150200 American41447001010770150195230260305350405 Society45747001010755105150180225255305 for4904700101073080115 Photogrammetry50457001010755110160190240285320365445525570600635685 and48907109010745100155 Remote50707109010760105185235265310 Sensing54057109010755100155195225280325 doi:45347237010755105135160 10.14358/PERS.80.3.239471972370107501001251752252753253754104655205806356607107607858368619119611011 ASPRS553187016183157240330403 MEMBERS9881870161108191299385468558631 AND1651187016183175267 SUBSCRIBERS,1949187016174165251324410499548634717807880911 CLICK2891187016186165214300390 HERE3312187016196179269352 TO3696187016179171 READ3899187016189173256348 THE4278187016180176259 ENTIRE4569187016183175254303393476 ARTICLE5077187016183173252301387466550