Building Damage Detection in Satellite Imagery Using Convolutional Neural Networks
Computer Vision and Pattern Recognition
2019-10-16 v1 Machine Learning
Image and Video Processing
Machine Learning
Abstract
In all types of disasters, from earthquakes to armed conflicts, aid workers need accurate and timely data such as damage to buildings and population displacement to mount an effective response. Remote sensing provides this data at an unprecedented scale, but extracting operationalizable information from satellite images is slow and labor-intensive. In this work, we use machine learning to automate the detection of building damage in satellite imagery. We compare the performance of four different convolutional neural network models in detecting damaged buildings in the 2010 Haiti earthquake. We also quantify how well the models will generalize to future disasters by training and testing models on different disaster events.
Cite
@article{arxiv.1910.06444,
title = {Building Damage Detection in Satellite Imagery Using Convolutional Neural Networks},
author = {Joseph Z. Xu and Wenhan Lu and Zebo Li and Pranav Khaitan and Valeriya Zaytseva},
journal= {arXiv preprint arXiv:1910.06444},
year = {2019}
}