Automated Detection of Damaged Areas after Hurricane Sandy using Aerial Color Images

Abstract

Rapid detection of damaged buildings after natural disasters, such as earthquakes and hurricanes, is an urgent need for first response, rescue and recovery planning. In this context, post-event aerial images which could be collected right after disasters are valuable sources for damage detection. However, manual analysis process of the acquired imagery could be both time consuming and costly. To address this issue, a series of classification models for post-hurricane automated detection of damaged buildings is presented in this paper. First, five feature sets were generated through feature extraction and transformation. Then, several classifiers were trained using two groups of classification methods: (1) the Minimum-distance and (2) the Support Vector Machine (SVM) methods. The effectiveness of these classifiers was evaluated in terms of classification accuracies and testing time. The results demonstrated the combination of feature sets and classification methods can provide the best performance. Furthermore, optimal classifiers were selected for future automated real-time damaged building detection. The observed performances of these optimal classifiers indicate promising application for a wide variety of image-based classification tasks.

Towards an Integrated Approach to Infrastructure Damage Assessment in the Aftermath of Natural Hazards

“…Ye et al [26] compared various approaches for the automated detection of damaged buildings using aerial imaging. Feature sets were firstly created from the images and then trained and tested via selected classification algorithms.…”

Visual monitoring of civil infrastructure systems via camera-equipped Unmanned Aerial Vehicles (UAVs): a review of related works

“…(Cho et al 2015;Pătrăucean et al 2015;Son et al 2015;Teizer 2015;Yang et al 2015) provide thorough reviews of these techniques. These all methods can be applied to visual data Table 1 The level of autonomy in UAV-based visual data collection for construction performance monitoring and civil infrastructure condition assessment None (Gheisari et al 2014;Irizarry et al 2012;Oskouie et al 2015;Vetrivel 2999;Wen et al 2014;Xie et al 2012;Ye et al 2014) collected from UAVs, yet only a few studies have validated them in such contexts. Tables 2 and 3 summarize the most recent literature from the last few years that focus on methods that are exclusively developed or applied to images and videos from UAVs.…”

“…Achieving effective flow of information both to and from project sites and conducting actionable analytics for construction monitoring and condition assessment require intuitive visualization of the information produced throughout the process on top of the UAV visual data - Fernandez Galarreta et al 2015); (Kerle 2999) • Machine learning-based classification of damaged buildings using feature sets obtained from feature extraction and transformation in images (Ye et al 2014) Infrastructure inspection • Image-based 3D reconstruction• Geometrical feature recognition and classification for planning laser scans (Oskouie et al 2015) • Creating comprehensive, high-resolution, semantically rich 3D models of infrastructure Geo-hazard investigations • Orthophoto mapping and visual interpretation to inspect geologic hazards along oil and gas pipelines (Gao et al 2011) images and point clouds. While attention to visual sensing and analytics has been the mainstream of the literature, less work is conducted on interactive visualization.…”