Detection the Urban Change Areas of Yangon City Using Landsat Time series Images

Abstract

Urban growth is the critical task for city planning of the developing country. It can estimate to know the increasing rate of the building area of the certain township during the specific year. The system proposes a method combining the Morphological Building Index (MBI) and Slow Feature Analysis (SFA). It can find the urban changing areas of the Yangon city using the Landsat 7 ETM+ time series images from 2003 to 2015. In MBI, it leads to a number of false alarms involving non-building urban structures such as soil and roads. In SFA, it alone is not suitable for building change detection since it provides high commission error. The purposed system combines these two method to overcome the weakness of MBI and SFA. The experimental result shows the comparative accuracies of MBI and SFA method only with the proposed method.