Using Deep Learning and Satellite Imagery to Assess the Damage to Civil Structures After Natural Disasters
Since 1980, millions of people have been harmed by natural disasters that have cost society over three trillion dollars. After a natural disaster has occurred, the creation of maps that identify the damage to buildings and infrastructure is imperative. Currently, many organizations perform this task manually, using pre- and post-disaster images and well-trained humans to infer the degree and extent of damage. This manual task can take days to complete. We propose to do this task automatically using post-disaster satellite imagery. We use a pre-trained neural network, SegNet, and replace its last layer with our own damage classification scheme. The final layer of the network is re-trained using cropped segments of the satellite image of the disaster. Our test results show that it is possible to create these maps quickly and efficiently.
Paper
Full text
Using Deep Learning and Satellite Imagery to Assess the Damage to Civil Structures After Natural Disasters
Semantic Scholar · Engineering · 2019
Abstract
Since 1980, millions of people have been harmed by natural disasters that have cost society over three trillion dollars. After a natural disaster has occurred, the creation of maps that identify the damage to buildings and infrastructure is imperative. Currently, many organizations perform this task manually, using pre- and post-disaster images and well-trained humans to infer the degree and extent of damage. This manual task can take days to complete. We propose to do this task automatically using post-disaster satellite imagery. We use a pre-trained neural network, SegNet, and replace its last layer with our own damage classification scheme. The final layer of the network is re-trained using cropped segments of the satellite image of the disaster. Our test results show that it is possible to create these maps quickly and efficiently.