Fully convolutional Siamese neural networks for buildings damage assessment from satellite images

TL;DR

This paper presents a Siamese neural network approach for automated damage assessment of buildings from satellite images before and after natural disasters, enabling efficient large-scale analysis.

Contribution

It introduces a novel Siamese encoder-decoder neural network architecture for damage classification, with extensive ablation studies and competitive results in a relevant challenge.

Findings

Achieved top performance in damage assessment competition

Compared various encoders, decoders, and loss functions

Demonstrated effective image comparison for damage classification

Abstract

Damage assessment after natural disasters is needed to distribute aid and forces to recovery from damage dealt optimally. This process involves acquiring satellite imagery for the region of interest, localization of buildings, and classification of the amount of damage caused by nature or urban factors to buildings. In case of natural disasters, this means processing many square kilometers of the area to judge whether a particular building had suffered from the damaging factors. In this work, we develop a computational approach for an automated comparison of the same region's satellite images before and after the disaster, and classify different levels of damage in buildings. Our solution is based on Siamese neural networks with encoder-decoder architecture. We include an extensive ablation study and compare different encoders, decoders, loss functions, augmentations, and several methods to combine two images. The solution achieved one of the best results in the Computer Vision for Building Damage Assessment competition.