Abstract:
People nowadays use social media platforms to capture and share real-time
incidents in the form of images, videos and text. However, sharing too much information at once makes it harder for first responders to determine where exactly
individuals are in need and whether they require immediate assistance. In the
past, machine learning techniques were used to automatically identify and infer
disaster response from images, as manually identifying disaster types is currently
challenging. Therefore, in this paper, deep learning models are used to investigate how well they can classify the images according to their disaster type by
learning the features extracted from the input images on their own. In this study,
2 existing datasets namely the ‘Comprehensive Disaster Dataset’ (CDD) and
‘Natural Disaster Dataset’ (NDD) based on disaster types were customized into
a dataset entitled as ‘Customized Disaster Dataset’. The Customized Disaster
Dataset comprises of a total of ten classes, three of which are non-damage images,
Pre-trained models like the MobileNetV2, VGG16 and InceptionV3 were used to
train the datasets to allow for further comparison with existing studies. Along
with that, a customized neural network model was created and trained on the
datasets. Different scenarios were devised to assess the top 3 performing models.
The InceptionV3 being best model had a classification accuracy of 96.86%. In
this study, we have demonstrated the effectiveness of CNN models as a tool for
automatic disaster type classification.
