Classification of IR Images of Small Eruptions at the Erebus Volcano, Antarctica, with Zernike moments and Support Vector Machine

This repository provides an "SVM_binary_classification" Jupyter notebook and different data folders for training and generalization in a binary classification task using a combination of support vector classifier and Zernike moments. Using a python script in the jupyter notebook and a Zernike moment file inside each folder, one can classify generalization images in any two classes among eruption, no-eruption, and non-explosive small events (NESE). It is notable that, the SVM needs to be trained initially using the Zernike moments inside the training data folders.

There is also spatial and temporal information on NESEs in December 2013, December 2014, and January 2015. reported in excel files.