01882nam a2200241   450000500170000000800410001703500150005810000170007324500700009026000520016030000350021249000270024750001050027450200580037952008890043765000360132665000410136270000320140370000440143570000370147981000590151685600650157520260817172817.0080998         th                  eng    a.b101412121 aLee, Ha-sook10aClassification technique for multi-temporal vegetation index data  aBangkok :bAsian Institute of Technology,c1993  a55 leaves +e1 online resource1 aThesis ;vno. CS-93-32  aA thesis submitted in partial fulfillment of the requirement for the degree of Master of Engineering  aThesis (M.Eng.) - Asian Institute of Technology, 1993  aGVI data were used in this digital image processing study to classify the land cover and to monitor the vegetation dynamics for Asian region. Since this data has multi-bands, in order to reduce the redundancy, Principal Component Analysis was implemented for this multi-temporal data. Color composite images of monthly GVI data were to be useful for visual interpretation of seasonal dynamics. There was a correspondence between seasonal and regional variation of the monthly GVI data. The pattern classification was applied to monthly data for one-year, using the Clustering which is useful for global land cover classification without ground truth. All proposed methods have been applied to GVI images of Asia, and good results have been obtained. Here it is proved that the multi-temporal data can be a suitable method to define the class type and to detect the vegetation changes. 0aRemote sensingxData processing 0aImage processingxDigital techniques1 aMurai, Shunji,eChairperson1 aPhan, Minh Dung,eExamination Committee1 aYulu, Qi,eExamination committee2 aAsian Institute of Technology.tThesis ;vno. CS-93-32  3Full-Textuhttp://203.159.5.9/ait-thesis/detail.php?q=B16142