Remote sensing-based detection and quantification of roadway debris following natural disasters
Remote sensing-based detection and quantification of roadway debris following natural disasters
复制标题
自然灾害后道路碎片的遥感检测和量化
DOI：
--
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
J. Holguín
中科院分区：
文献类型：
--
作者：
Colin Axel;J. V. van Aardt;Felipe Aros;J. Holguín
快速了解道路网络状况对于在任何重大灾害发生后制定有效的应急响应计划至关重要。倒塌的建筑物、无法移动的车辆和其他形式的碎片往往使道路无法通行。道路状况通常是通过时间和资源密集的方法确定的，例如实地调查和人工判读遥感图像。机载激光雷达系统为执行网络评估提供了一种替代的、具有成本效益的选择。3D数据可以在大范围内快速收集，并提供有关场景几何和结构的宝贵见解。提出了一种利用机载激光雷达数据自动检测和表征道路中碎片的方法。落在道路范围内的点从点云中提取，并使用区域生长聚类成单个对象。根据物体的表面特性和上下文线索，将物体分为碎片和非碎片。碎片堆被重建为使用阿尔法形状的表面，从中可以计算碎片量的估计。结果使用真实的激光雷达数据收集后的自然灾害。初步结果表明，准确的碎片地图可以自动生成使用所提出的方法。这些碎片地图将是一个宝贵的资产，灾害管理和应急小组试图接触幸存者，尽管交通网络瘫痪。
Rapid knowledge of road network conditions is vital to formulate an efficient emergency response plan following any major disaster. Fallen buildings, immobile vehicles, and other forms of debris often render roads impassable to responders. The status of roadways is generally determined through time and resource heavy methods, such as field surveys and manual interpretation of remotely sensed imagery. Airborne lidar systems provide an alternative, cost-e↵ective option for performing network assessments. The 3D data can be collected quickly over a wide area and provide valuable insight about the geometry and structure of the scene. This paper presents a method for automatically detecting and characterizing debris in roadways using airborne lidar data. Points falling within the road extent are extracted from the point cloud and clustered into individual objects using region growing. Objects are classified as debris or non-debris using surface properties and contextual cues. Debris piles are reconstructed as surfaces using alpha shapes, from which an estimate of debris volume can be computed. Results using real lidar data collected after a natural disaster are presented. Initial results indicate that accurate debris maps can be automatically generated using the proposed method. These debris maps would be an invaluable asset to disaster management and emergency response teams attempting to reach survivors despite a crippled transportation network.