DOI: 10.1016/j.jag.2014.06.011
Scopus记录号: 2-s2.0-84904766959
论文题名: Enhancing endmember selection in multiple endmember spectral mixture analysis (MESMA) for urban impervious surface area mapping using spectral angle and spectral distance parameters
作者: Fan F ; , Deng Y
刊名: International Journal of Applied Earth Observation and Geoinformation
ISSN: 15698432
出版年: 2014
卷: 33, 期: 1 起始页码: 290
结束页码: 301
语种: 英语
英文关键词: Endmember selection
; Impervious surface area
; MESMA
; Spectral angle
; Spectral distance
Scopus关键词: image analysis
; mapping
; pixel
; remote sensing
; spectral analysis
; urban area
; China
; Guangdong
; Guangzhou
英文摘要: Successful retrieval of urban impervious surface area is achieved with remote sensing data using the multiple endmember spectral mixture analysis (MESMA). MESMA is well suited for studying the urban impervious surface area because it allows the number and types of the endmembers to vary on a per-pixel basis, thereby, allowing the control of the large spectral variability. However, MESMA must calculate all potential endmember combinations of each pixel to determine the best-fit one. Therefore, it is a time-consuming and inefficient unmixing technology, especially for hyperspectral images because these images have more complicated endmember categories. Hence, in this paper, we design an improved MESMA (SASD-MESMA: spectral angle and spectral distance MESMA) to enhance the computational efficiency of conventional MESMA, and we validate this new method by analyzing the Hyperion image(Jan-2011) and the field-spectra data of Guangzhou (China). In SASD-MESMA, the parameters of spectral angle (SA) and spectral distance (SD) are used to evaluate the similarity degree between library spectra and image spectra in order to identify the most representative endmember combination for each pixel. Results demonstrate that the SA and SD parameters are useful to reduce misjudgment in selecting candidate endmembers and effective for determining the appropriate endmembers in one pixel. Meanwhile, this research indicates that the proposed SASD-MESMA performs very well in retrieving impervious surface area, forest, grass and soil distributions on the sub-pixel level (the overall root mean square error (RMSE) is 0.15 and the correlation coefficient of determination (R2) is 0.68). © 2014 Elsevier B.V. Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/79760
Appears in Collections: 气候变化事实与影响
There are no files associated with this item.
作者单位: School of Geography, South China Normal University, Guangzhou 510631, China; Department of Geography, University of Wisconsin-Milwaukee, Milwaukee, WI 53201-0413, United States
Recommended Citation:
Fan F,, Deng Y. Enhancing endmember selection in multiple endmember spectral mixture analysis (MESMA) for urban impervious surface area mapping using spectral angle and spectral distance parameters[J]. International Journal of Applied Earth Observation and Geoinformation,2014-01-01,33(1)