Detection of Urban Expansion by using DMSP-OLS, Landsat Data and Linear Spectral Unmixing Method

Detection of Urban Expansion by using DMSP-OLS, Landsat Data and Linear Spectral Unmixing Method

Cihan Uysal (Department of Geomatics Engineering, Istanbul Technical University, Istanbul, Turkey) and Derya Maktav (Department of Geomatics Engineering, Istanbul Technical University, Istanbul, Turkey)
Copyright: © 2015 |Pages: 10
DOI: 10.4018/IJ3DIM.2015040105
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Urbanization has been increasingly continuing in Turkey and in the world for the last 30 years. Especially for the developing countries, urbanization is a necessary fact for the sustainability of the urban growth. Yet, this growth should be controlled and planned; otherwise, many environmental problems might occur. Therefore, the urban areas having dynamic structure should be monitored periodically. Monitoring the changes in urban environment can be provided with land cover land use (LCLU) maps produced by the pixel based classification methods using ‘maximum likelihood' and ‘isodata' techniques. However, these thematic maps might bring about inaccurate classification results in heterogeneous areas especially where low spatial resolution satellite data is used since, in these approaches, each pixel is represented with only one class value. In this study, considering the spectral mixture analysis (SMA) each pixel is represented by endmember fractions. The earth is represented more accurately using 'substrate (S)', ‘green vegetation (V)' and ‘dark surfaces (D)' spectral endmember reflectances with this analysis based on linear mixture model. Here, the surrounding of Izmit Gulf, one of the most industrialized areas of Turkey, has been chosen as the study area. SMA has been applied to LANDSAT images of the years of 1984, 1999 and 2009. In addition, DMSP-OLS data of 1992, 1999 and 2009 has been used to detect urban areas. According to the results, the changes in LCLU and especially the urban growth areas have been detected accurately using the SMA method.
Article Preview

Study Area And Data

In the study, Izmit Gulf and its surroundings which is situated in the east of the Sea of Marmara in Marmara Region in Turkey have been selected as the study area (Figure 1). The city of Izmit lies on latitude 400 41’ and longitude 290 31.

Figure 1.

The study area

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 6: 4 Issues (2017): 2 Released, 2 Forthcoming
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2014)
Volume 2: 4 Issues (2013)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing