Change Detection of the Building Areas in Urban regions

Abstract

Myanmar is exposed to a number of natural hazards, some of which have caused devastating damage in the recent past. According to the UN Risk Model, Myanmar ranks as the ‘most at risk’ country for natural disasters. Coastal regions, particularly Rakhine State and the Ayeyarwady Delta Region, are at high risk for cyclones, storm surges and tsunamis. As Myanmar falls on one of the two main earthquake belts in the world, much of the country is prone to earthquake. After the disaster occurred, we cannot classify completely the building areas on widely damaged urban areas many years later. Urban growth is the critical feeds for the city planning and directly effects on the country development. The modified Morphological Building Index (MBI) is applied to extract building areas to know how much area has change. Then, matching-based change rule is applied to obtain changes areas. In the experiments show that the proposed method can achieve satisfactory correctness rate by comparing with Change Vector Analysis (CVA).