000 03877nas|a2200457 i 4500
005 20260817170903.0
008 030530s2002 th u m rtt 0| a1eng d
035 _a.b11885038
099 9 _aAIT Thesis no.SR-02-09
100 0 _aTrinh Thi Phin
245 1 0 _aAutomated damage detection of buildings from high resolution satellite images case study :
_bthe 2001 Gujarat, India earthquake
260 _aBangkok :
_bAsian Institute of Technology,
_c2002
300 _a58 leaves
490 1 _aThesis ;
_vno. SR-02-09
502 _aThesis (M. Eng.) - Asian Institute of Technology, 2002
500 _aA thesis submitted in partial fulfillment of requirements for the degree of Master of Engineering, School of Advanced Technologies
520 _aMapping of the distribution damaged building after earthquake is important to know where is the building that collapsed. Earthquake events are usually of short duration. It is very important to obtain real-time information of building damage in the urban area in order to rescue the people that trapped on the building. In this study, a maximum likelihood classification image is used. An automatic detection method is built for real time monitoring. This method produces distribution map as final output from high- resolution satellite image. Once a high-resolution image obtained after the event, a final output is provided concurrently with existing building damaged situation. At the first stage, a single satellite image is classified to extirpate the extractable classes. Principal component analysis is applied on the image after removing the extractable classes. New image is created by band mathematical operation, which is the division of the first component image to the all band of the image after removing the extractable classes. At the second stage, this new image is classified with the same position of the training data but the classes that correspond to the extractable are pruned. Then, the building damaged level will be presented on the original image. Through this simulation method, damaged buildings are automatically mapped as the presentation of real-time distribution. This method inherits the classification of land cover. This method presents good results for the urban area classes. In particular, the class denoting collapsed buildings agreed well with the actual situation of damaged building, and the application of this method to real-time earthquake disaster management can be expected.
650 0 _aBuildings
_xEarthquake effects
_zIndia
650 0 _aRemote sensing images
_zIndia
700 1 _aYamazaki, Fumio,
_eChairperson
700 1 _aYokoyama, Ryuzo,
_eExamination committee
700 1 _aKusanagi, Michiro,
_eExamination Committee
700 1 _aBorne, Frederic,
_eExamination committee
710 2 _aASEAN Foundation,
_eScholarship donor
810 2 _aAsian Institute of Technology.
_tThesis ;
_vno. SR-02-09
856 _3Full-Text
_uhttp://203.159.5.9/ait-thesis/detail.php?q=B07519
907 _a.b11885038
_bmnait
_cu
902 _a240403
998 _b0
_c030530
_dm
_ea
_fu
_g0
962 _a000:001:PDF:b1188503:002020:0:0:0:0:0:0
_tAbstract--AIT Thesis no.SR-02-9
_vn
945 _lmnait
945 _lmnait
945 _lmnarc
942 _c22
942 _c20
942 _c40
909 _aBarcode : 30050120694285
_bCREATED : 2014-04-28
_cRECORD # : i1279269x
_dLPATRON : 0
_eLCHKIN : -
_f# RENEWALS : 0
_g# OVERDUE : 0
_hIUSE3 : 0
_iTOT CHKOUT : 0
_jTOT RENEW : 0
909 _aBarcode : 30050120694319
_bCREATED : 2015-12-02
_cRECORD # : i12839243
_dLPATRON : 1025945
_eLCHKIN : 2015-06-20
_f# RENEWALS : 0
_g# OVERDUE : 0
_hIUSE3 : 0
_iTOT CHKOUT : 1
_jTOT RENEW : 0
909 _aBarcode : 30050160022892
_bCREATED : 2016-01-04
_cRECORD # : i1291969x
_dLPATRON : 0
_eLCHKIN : -
_f# RENEWALS : 0
_g# OVERDUE : 0
_hIUSE3 : 0
_iTOT CHKOUT : 0
_jTOT RENEW : 0
999 _c10123
_d10123