Comparison of Principal Components Analysis and Minimum Noise Fraction Transformation for Reducing the Dimensionality of Hyperspectral Imagery

dc.contributor.author陳哲銘zh_tw
dc.date.accessioned2014-10-27T15:40:27Z
dc.date.available2014-10-27T15:40:27Z
dc.date.issued2000-11-??zh_TW
dc.description.abstract超光譜影像通常具有兩百個以上的連續波段,且各波段間存在高度相關,因此在影像分析前常需減少資料量以提昇運算效率,同時消除波段間的相關以減少分析誤差。本研究比較主成分分析與最小雜訊區段轉換這兩種轉換方法對減少超光譜影像資料維數的成效,結果證實最小雜訊區段轉換可正確根據影像品質的高低來排列主成分的次序,且有較高的訊號雜訊比,因此比主成分分析更適用於壓縮超光譜資料。zh_tw
dc.identifier601A45EF-D21E-8AAA-DFED-1B5D523AC031zh_TW
dc.identifier.urihttp://rportal.lib.ntnu.edu.tw/handle/20.500.12235/23692
dc.language英文zh_TW
dc.publisher地理學系zh_tw
dc.publisherDepartment of Geography, NTNUen_US
dc.relation(33),163-178zh_TW
dc.relation.ispartof國立臺灣師範大學地理研究報告zh_tw
dc.subject.other超光譜影像zh_tw
dc.subject.other資料維數zh_tw
dc.subject.other主成分分析zh_tw
dc.subject.other最小雜訊區段轉換zh_tw
dc.subject.other訊號雜訊比zh_tw
dc.titleComparison of Principal Components Analysis and Minimum Noise Fraction Transformation for Reducing the Dimensionality of Hyperspectral Imageryzh-tw
dc.title.alternative主要成分分析與最小雜訊區段轉換對減少超光譜影像維數之比較zh_tw

Files

Collections