Deep Neural Networks are used to extract information from images. The Relevance, Type of Damage and the Severity is identified.

Since images in the extracted posts may be used in contexts unrelated to the disaster taking place, the first stage is to discard irrelevant images.
Try NowWe use Deep Learning models to classify the relevant images into fire damage, natural damage, infrastructure damage, and flood damage.
Try Now

The damage depicted in the images is classified as mild and severe to assist rescue operations.
Try Now