Please use this identifier to cite or link to this item:
標題: 利用空間及頻率濾波器降噪以提升高光譜影像分類
Improving hyperspectral image classification through image denoising by spatial and frequency fliters
作者: 王奕翔
Wang, Yi-Hsiang
關鍵字: 降低雜訊;denoise;高光譜影像;分類準確度;MNF;支持向量機;外埔;hyperspectral;Minimum Noise Fraction;Support Vector Machine
出版社: 土木工程學系所
引用: 1.黃凱翔,2008,利用支持向量機於機載高光譜感測影像之分類,國立中興大學土木工程所碩士論文。 2.黃弘毅,2012,萃取方式於ISIS高光譜影像分類判釋之比較,國立中興大學土木工程所碩士論文。 3.陳繼藩、徐守道、陳世旺,1997,應用非監督性類神經網路於SPOT衛星影像分類之研究,航測及遙測學刊,2(1),pp.1-12。 4.王儷蓉,1998,高光譜資料之差異與類別分離度分析,國立成功大學 測量工程研究所碩士論文。 5.郭麟霂,2000,寒帶沼地高光譜影像分類之研究,國立交通大學土木工程學系碩士論文。 6.楊士興,2001,應用獨立成分分析方法分離集水區河川污染源之研究,國立台灣大學地理環境資源學研究所博士論文。 7.林金樹,2002,高光譜主軸轉換影像辨識土地利用型最適主軸數決定法之研究,臺灣林業科學,17(4),pp.471-481。 8.魏曉萍,2003,QuickBird衛星影像探討分類方法之研究,中華大學土木工程學系碩士論文。 9.陳建州,2004,利用獨立成分分析法在區域特徵上的人臉辨識,成功大學資訊工程學系碩士論文。 10.林恩楷,2005,利用高光譜影像偵測外來植物-以恆春地區銀合歡為例,中央大學土木工程研究所碩士論文。 11.高莉,2007,結合ICA和PCA的腦磁信號消噪研究。 12.蕭國鑫,2008,高解析度影像應用於土地利用分類之探討,航測及遙測學刊,13(4)。 13.蔡富安,2007,遙測影像條紋雜訊改正之研究,國立中央大學太空及遙測研究中心 14.曾義星、賴志恆,2003,應用遙測於台北都會區綠地環境變遷分析,航測及遙測學刊,12(2),pp.145-163。 15.雷祖強、周天穎、萬絢、楊龍士、許晉嘉,2007,空間特徵分類器支援向量機之研究,航測及遙測學刊,8(1),pp.35-46。 16.Amato, U., and Cavalli, R., 2009. "Experimental Approach to the Selection of the Components in the Minimum Noise Fraction", IEEE Transactions on Geoscience and Remote Sensing, 47(1):153-160. 17.Albani, M., and Klinkenberg, B., 2003. "A Spatial Filter for the Removal of Striping Artifacts in Digital Elevation Models", Photogrammetric Engineering & Remote Sensing, 69(7):755-765. 18.Chang, C.I., 2006. "Recent Advances in Hyperspectral Signal and Image Processing", Transworld Research Network. 19.Chang, C.I., Zhao, X.L., ang Althouse, L.G., 1999. "Least Squares Subspace Projection Approach to Mixed Pixel Classification for Hyperspectral Images", IEEE Transactions on Geoscience and Remote Sensing, 36(3):898-912. 20.Chang, C.I., 2006. "Recent Advances in Hyperspectral Signal and Image Processing", Transworld Research Network. 21.Horn, B.K.P., and Woodham, R.J., 1979. "Destriping Landsat MSS Imagery by Histogram Modification", Computer Graphics and Image Processing, 10:68-83. 22.Lillesand, T.M., Kiefer, R.W., and Chipman, J.W., 2004. "Remote Sensing and Image Interpretation", 5th ed., John Wiley & Sons, New York, USA. 23.Landgrebe, D., 2002. "Hyperspectral Image Data Analysis", IEEE Signal Processing Magazine, 19(1):17-28. 24.Liu, J.G. and Gareth L.K. Morgan, 2006. "FFT Selective and Adaptive Filtering for Removal of Systematic Noise in ETM+ Imageodesy Images", IEEE Transactions on Geoscience and Remote Sensing(accepted). 25.Lee, J.S., "Digital Image Smoothing and The Sigma Filter", Computer Vision, Graph. Image Process, 24(2):255. 26.Nill, N., and Bouzas, B., 1992. "Objective Image Quality Measure Derived from Digital Image Power Spectra", Optical Engineering, 3(4):813-825. 27.Oimoen, M.J., 2000. "An Effective Filter for Removal of Production Artifacts in U.S. Geological Survey 7.5-minute Digital Elevation Models", Proc. of the Fourteenth International Conference on Applied Geologic Remote Sensing, Las Vegas, NUSAV, pp311-319. 28.Rakwatin, P., Takeuchi, W., and Yasuoka, Y., 2005. "Destriping MODIS Data by Facet Model and Histogram Matching", Proc. 26th Asian Conference on Remote Sensing (ACRS 2005), Hanoi, Vietname, CD-ROM. 29.Switzer, P., and Green, A.A., 1984. "Min /Max Autocorrelation Factors for Multivariate Spatial Imaging", Technical Report No.6, Department of Statistics, Stanford University, 14p. 30.Truong, T.K., Wang, L.J., Reed, I.S., Hsieh, W.S., 2000. "Image Data Compression Using Cubic Convolution Spline Interpolation", IEEE Transactions Image Processing, 9(11):1988-1995. 31.Tsai, F., and Philpot, W. D., 2002. "A Derivative-aided Hyperspectral Image Analysis System for Land-cover Classification", IEEE Transactions on Geoscience and Remote Sensing, 40(2):416-425. 32.Unser, M., 1999. "Splines: A Perfect Fit for Signal and Image Processing", IEEE Signal Processing, 16(6):22-38. 33.van Puymbroeck, N., Michel, R., Binet, R., Avouac, J. P., and Taboury, J., 2000. "Measuring Earthquakes from Optical Satellite Images", Applied Optics, 39(20): 3486–3494.

 Hyperspectral image provides more information and has better spectral resolution. The Waipu testsite is established by The Center of Environmental Restoration and Disaster Reduction at National Chung Hsing University, and the test field has 200 ha of rice paddy for a long term monitoring.
  The ISIS hyperspectral imager made by National Applied Research Laboratories contains 218 bands with a spectral bandwidth of 3-5 nm between 427.2 nm and 945.7 nm. ISIS is a push-broom scanner with 218 bands in each scan. Feature extraction can be done through Minimum Noise Fraction (MNF) regarded as the most effective approaches to solving the inefficiency of computation. This research applied Support Vector Machine (SVM) to do the classification.
  To dissolve the problem of scattered pixels and noises along the flying track on ISIS hyperspectral image, both frequency and spatial fliters were applied to reduce the noises. And spline-based interpolation is also applied in this research to improve the classification accuracy.
This research applies Minimum Noise Fraction (MNF), Fast Fourier Transform(FFT), Spline and Support Vector Machines(SVMs) to classification. According to the result of classification, the noise contains in ISIS hyperspectral image can be removed effectively. The overall accuracy of ISIS hyperspectral image can be promoted from 49% to 78.8%.
其他識別: U0005-2108201316371400
Appears in Collections:土木工程學系所

Show full item record
TAIR Related Article

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.