Landsat Images Classification and Change Analysis of Land Cover/Use in Istanbul

Authors : İsmail BÜYÜKSALİH

Pages : 56-65

Doi:10.30897/ijegeo.304484

View : 11 | Download : 6

Publication Date : 2016-08-02

Article Type : Research Paper

Abstract :This paper describes the methodology and results of classifications of Landsat TM data of the Istanbul for the years 1987 and 2007. Nine different land cover/use categories have been used, named built-up area, cropland, barren ground, grassland, scrub/brush, water, ever green, deciduous, cloud and others uses. When the obtained classification results are evaluated as a result of, the 1987 Landsat image overall accuracy of 79 % and a kappa value of 0.76, the 2007 Landsat image overall accuracy of 83.50% and kappa value of 0.81. Thus, Istanbul`s change analysis was revealed that the 20-year period. The classifications have provided an economical and accurate way to quantify, map and analyze changes over time in land cover. 
Keywords : Remote sensing, Land cover Land use, GIS, Temporal Analysis, Landsat