Share Email Print
cover

Proceedings Paper

Assessing and monitoring of urban vegetation using multiple endmember spectral mixture analysis
Author(s): M. A. Zoran; R. S. Savastru; D. M. Savastru
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

During last years urban vegetation with significant health, biological and economical values had experienced dramatic changes due to urbanization and human activities in the metropolitan area of Bucharest in Romania. We investigated the utility of remote sensing approaches of multiple endmember spectral mixture analysis (MESMA) applied to IKONOS and Landsat TM/ETM satellite data for estimating fractional cover of urban/periurban forest, parks, agricultural vegetation areas. Because of the spectral heterogeneity of same physical features of urban vegetation increases with the increase of image resolution, the traditional spectral information-based statistical method may not be useful to classify land cover dynamics from high resolution imageries like IKONOS. So we used hierarchy tree classification method in classification and MESMA for vegetation land cover dynamics assessment based on available IKONOS high-resolution imagery of Bucharest town. This study employs thirty two endmembers and six hundred and sixty spectral models to identify all Earth’s features (vegetation, water, soil, impervious) and shade in the Bucharest area. The mean RMS error for the selected vegetation land cover classes range from 0.0027 to 0.018. The Pearson correlation between the fraction outputs from MESMA and reference data from all IKONOS images 1m panchromatic resolution data for urban/periurban vegetation were ranging in the domain 0.7048 - 0.8287. The framework in this study can be applied to other urban vegetation areas in Romania.

Paper Details

Date Published: 5 August 2013
PDF: 9 pages
Proc. SPIE 8795, First International Conference on Remote Sensing and Geoinformation of the Environment (RSCy2013), 87951Y (5 August 2013); doi: 10.1117/12.2027500
Show Author Affiliations
M. A. Zoran, National Institute of Research and Development for Optoelectronics (Romania)
R. S. Savastru, National Institute of Research and Development for Optoelectronics (Romania)
D. M. Savastru, National Institute of Research and Development for Optoelectronics (Romania)


Published in SPIE Proceedings Vol. 8795:
First International Conference on Remote Sensing and Geoinformation of the Environment (RSCy2013)
Diofantos G. Hadjimitsis; Kyriacos Themistocleous; Silas Michaelides; George Papadavid, Editor(s)

© SPIE. Terms of Use
Back to Top