Object-oriented Remote Sensing Data Classification Method Application in Earthquake Damage Rapid Extraction
WANG Yan;WANG Xiao-qing;DOU Ai-xia;SUN Su-mei;FU Cong;Earthquake Administration of Liaoning Province;Institute of Earthquake Forecasting,CEA;
Beichuan County and its surrounding area where were severely damaged were chosen as the study area. And the object-oriented remote sensing data classification method was used to extract the damage area from the multi-spectral data of Formosat-2. During the extraction, we added geometric structure and texture information into the spectral information was the only information used in the pixel-based classification, and applied multi-resolution segmentation and multi-level crossing mode to simulate visual interpretation, and adopted fuzzy-logic classification method to accomplish the effective detection.