This repository is part of my bachelor´s degree project. This is focused on the analysis and enhancement of all the processes related to the generation and implementation of a Convolutional Neural Network.
Through this technique belonging to the field of Deep Learning, it has been possible to design an application which allows us to detect whether there are damaged zones for an earthquake impact in an image. Therefore, it is used a specific topology for this problem. Specifically, the images used are provided by the earthquake occurred in Haití in 2010, obtained from an earth observation satellite called GeoEye-1.
As an example, the images look like this:
Being the main goal to implement this neural network in Intel´s devices. Hence, we will use the OpenVino Toolkit to obtain a multiplatform and optimized model to be used in Intel´s hardware.
This CNN (Convolutional Neural Network) is deployed in Keras by Tensorflow using Python. The version of Tensorflow is 2.2 for development, but for OpenVino, the v2.0 is needed (check the rest of tools and dependencies in OpenVino´s website). This model is formed by the next layers:
This is how the repository is structured:
This work is based on a project developed by K. J. Somaiya College of Engineering, Vidyavihar, Mumbai & Centre of Studies in Resources Engineering, IIT Bombay, Powai, Mumbai (India). This is the paper that I´ve been following.