Middle-East Journal of Scientific Research 13 (8): 1057-1064, 2013
ISSN 1990-9233
© IDOSI Publications, 2013
DOI: 10.5829/idosi.mejsr.2013.13.8.519
Corresponding Author: H. Matinfar, Department of Soil Science, Lorestan University, Lorestan, Iran.
1057
Decision Tree Land Use/ Land Cover Change Detection of
Khoram Abad City Using Landsat Imagery and Ancillary SRTM Data
Hamid Reza Matinfar and Majid Shadman Roodposhti
1 2
Department of Soil Science, Lorestan University, Lorestan, Iran
1
Department of Geography, Tehran University, Tehran, Iran
2
Abstract: Change detection is a general remote sensing technique that compares imagery collected over the
same area at different times and highlights features that have changed. In this paper, land cover of Khoram
Abad, a city in Lorestan province of Iran, was examined in a case study via post classification technique and
decision tree classifier. The Decision Tree (DT) classifier performs multistage classifications by using a series
of binary decisions to place pixels into proper classes. Input data may be used from various sources and data
types. Such as, multispectral data, digital elevation model (DEM) and slop to find features with similar spectral
reflectance but different in elevation. In order to carry out comprehensive analysis of Khoram Abad land cover
changes from years 1992 to 2009, TM data obtained from Landsat Satellite and digital elevation model of shuttle
radar topography mission were used. Finally, post classification analysis using DT classifier showed notable
improvement in classification accuracy in spite of high correlation of multi-spectral data.
Key words: Land Cover Post Classification Decision Tree Multi-spectral Images Change Detection
Khoram Abad Iran
INTRODUCTION detect land cover changes in Aykitali, Turkey. He found
The geospatial phenomena are changing over time among these techniques while PCA and post
and the land cover information has to be up-date classification analyses showed better results in change
periodically. Up-to-date knowledge of land cover is an detection.
important tool for the various planning authorities with Tardi and Contalgon [5] also used three methods
responsibilities for the management of territory [1]. including: multi-temporal color composite, subtraction and
However, it should be noted that planners and land classification in order to examine physical development of
managers require accurate data to address land cover Massachusett's urban area and resulting land cover
problems. Although the priority is for land use (economic) changes. Finally, they used post classification analysis in
information, land cover information is more easily mapped order to estimate total accuracy. Qiasvand [6], also
and can serve as an approximation of land use. concluded similar study via PCA and subtraction
During the past decades, not only remote sensing techniques so as to present south Tehran land cover map
images have become an important tool for land use and he reported that regression analysis in conjunction
classification and mapping [2] but also because of the with PCA showed better results. Jahani [7], utilized
advantages of repetitive data acquisition, they have satellite images (Spot) and normalized difference
become major data sources from local to global scales for vegetation index in Tehran land cover mapping project.
different change detection applications [3]. There were Consequently, on the basis of the earlier studies about
several studies conducted to investigate land use land cover change detection, it is obvious that most
changes during the time some of which will be referred researchers used subtraction and PCA techniques to
to in this section. Sunar [4], used five techniques, detect changes in land cover and, in further step; by
including: adding, subtracting, dividing, principle classifying multi-temporal images they showed results in
component (PCA) and post classification analysis to quantitative form.
that adding and subtracting images were the most simple