Improvement of linear spectral mixture analysis and experimentation in estimation of urban vegetation fraction

被引:0
|
作者
Yue, WZ [1 ]
Xu, JH [1 ]
Wu, JW [1 ]
机构
[1] Zhejiang Univ, Coll SE Land Management, Hangzhou 310029, Peoples R China
来源
IGARSS 2005: IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, VOLS 1-8, PROCEEDINGS | 2005年
关键词
D O I
暂无
中图分类号
P [天文学、地球科学];
学科分类号
07 ;
摘要
Abundance of vegetation plays an important role in urban ecosystem, urban planning and development. Traditional classification methods on remote sensing data by assigning each pixel membership in one, and only one have the primary shortcomings of their inability to accommodate spectrally mixed pixels in gradational land covers. The traditional classification methods are giving way to spectral mixture analysis (SMA) gradually which is better in acquiring quantitative information for specific hand covers. Vegetation fraction, in it general way, is defined as the areal fractions of vegetation within each pixel. This paper, besides introducing the traditional technique of SMA, discusses the improvement of traditional technique from the aspects of data noise removal, least-squares solution with constraining stun of endmembers fractions to unit, pixel purity index and the selection of endmembers. LSMA is tested further with the Shanghai city as an example. Unmixing pixels with root mean square (RMS) error less than 0.02 accounts for the proportion of 98.5% The spatial distribution of vegetation is corresponding to actual situation. Then we conclude that: the improved LSMA is appropriate for estimating quantitative vegetation fraction and the technique will be widely applied in urban environment.
引用
收藏
页码:1479 / 1482
页数:4
相关论文
共 50 条
  • [31] Kernel-Based Linear Spectral Mixture Analysis
    Liu, Keng-Hao
    Wong, Englin
    Du, Eliza Yingzi
    Chen, Clayton Chi-Chang
    Chang, Chein-I
    IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2012, 9 (01) : 129 - 133
  • [32] Linear spectral random mixture analysis for hyperspectral imagery
    Chang, CI
    Chiang, SS
    Smith, JA
    Ginsberg, IW
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2002, 40 (02): : 375 - 392
  • [33] Finding Virtual Signatures for Linear Spectral Mixture Analysis
    Song, Meiping
    Chen, Shih-Yu
    Li, Hsiao-Chi
    Chen, Hsian-Min
    Chen, Clayton Chi-Chang
    Chang, Chein-I
    IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2015, 8 (06) : 2704 - 2719
  • [34] Unsupervised linear spectral mixture analysis with AVIRIS data
    谷延锋
    杨冬云
    张晔
    Journal of Harbin Institute of Technology, 2005, (05) : 5 - 10
  • [35] INTER-ALGORITHM RELATIONSHIPS FOR RETRIEVALS OF FRACTION OF VEGETATION COVER IN A FRAMEWORK OF LINEAR MIXTURE MODEL
    Obata, Kenta
    Yoshioka, Hiroki
    NETWORKING THE WORLD WITH REMOTE SENSING, 2010, 38 : 859 - 864
  • [36] Fractional Vegetation Cover Estimation Based on an Improved Selective Endmember Spectral Mixture Model
    Li, Ying
    Wang, Hong
    Li, Xiao Bing
    PLOS ONE, 2015, 10 (04):
  • [37] Relative spectral mixture analysis - A multitemporal index of total vegetation cover
    Okin, Gregory S.
    REMOTE SENSING OF ENVIRONMENT, 2007, 106 (04) : 467 - 479
  • [38] Remote sensing of urban vegetation life form by spectral mixture analysis of high-resolution IKONOS satellite images
    Nichol, J.
    Wong, M. S.
    INTERNATIONAL JOURNAL OF REMOTE SENSING, 2007, 28 (05) : 985 - 1000
  • [39] Estimation of impervious surfaces of Beijing, China, with spectral normalized images using linear spectral mixture analysis and artificial neural network
    Hu, Xuefei
    Weng, Qihao
    GEOCARTO INTERNATIONAL, 2010, 25 (03) : 231 - 253
  • [40] Mapping Urban Impervious Surfaces by Using Spectral Mixture Analysis and Spectral Indices
    Li, Wenliang
    REMOTE SENSING, 2020, 12 (01)