ENIVRONMENT ASIA

Volume 12, No. 02, Month MAY, Year 2019, Pages 145 - 154

Land-cover change detection in a part of cameron highlands, malaysia using etm+ satellite imagery and support vector machine (svm) algorithm

Ayub Mohammadi, Himan Shahabi, and Baharin Bin Ahmad

Abstract Download PDF

The importance of the land cover map in many areas including environmental issues, regional planning, sustainable development and many other issues has been considered in the global literature. This study aims to detect land cover changes that occurred in a part of Cameron Highlands, Pahang, Malaysia for a decade between early 2008 until the late 2017. However, in this research Landsat-7 (ETM+) satellite data (2008 and 2017) were acquired and corrected using environment for visualizing images (ENVI) software. Support vector machine (SVM) model was used to classify land covers of both years. Results show that during this period of time forest has underwent considerable changes among all from 49.517532 km2 in 2008 to just 34.90341 km2 by the year 2017. The second substantial change belongs to vegetation and florification from 31.265611 km2 in 2008 to a peak of 42.83176 km2 by the year 2017. It means that forest mostly have been replaced by vegetation and florification. Furthermore, the overall accuracy and Kappa coefficient using confusion matrix for the year 2008, using support vector machine classification model are 96.4051% and 0.9294, respectively. While these values for the year 2017 are 97.5198% and 0.9674 in order. The current research can be valuable for land cover management in Cameron Highlands by Malaysian policy and decision makers.

Keywords

Remote sensing techniques; Geographical information system; Landsat image; Cameron Highlands

---