【摘要】 发生在四川省汶川县的"5.12"大地震给四川乃至全国带来了巨大的损失,道路是本次抗震救灾过程中的"生命线",如何对道路的损毁状况进行快速准确的监测成为抗震救灾的一项重要工作。以汶川地震的重灾区——北川县为研究对象,利用震前震后的高分辨率遥感影像,通过面向对象的图像分类方法进行道路识别,并通过震前道路识别结果与震后影像的叠加和震前震后道路识别结果的变化检测提取出损毁的路段。信息提取的结果已在第一时间提交给相关部门,为生命线的尽快抢通提供了及时、科学的依据。更多还原

【Abstract】 The earthquake occurred in the county of Wenchuan,Sichuan province on May 12th,2008 has brought huge loss in both property and lives.Road is the lifeline which is very important for disaster relief and people’s life after the earthquake.Thus,it becomes one of the most significant tasks for monitoring the situation of road damage.This study selects Beichuan which was destroyed seriously in the earthquake as the study area.Utilizing high spatial resolution remote sensing images before and after the earthquake,this study recognizes roads based on object-oriented image classification technology.In addition,it extracts the sections of damaged road using two approaches,one of which is overlaying the roads extracted from images acquired before the quake with the images after the quake,while the other is applying change detection to the roads extracted from images before and after the quake.Result of this study has already submitted to correlative departments of ministry of communications and provided timely and scientific basis for the disaster relief.This pilot study shows the methods mentioned above is feasible to road damage extraction based on high-resolution airborne image and satellite images.However,further study should be done in the near future.更多还原