challenge 7-Lift The Veil project

Inspiration

When driving in bad weather conditions, train drivers often cannot see signals and track-side equipment in front of their engine making the planning of the journey innacurate and increasing the probability of delays. This inability of seeing the outside is not just about drivers slowing down the engine but passengers also missing the fun of enjoying amazing landscapes.

We understand this can be frustrating based on our own experience. As a group of technology enthusiasts we believe in the power of machine learning and data-driven decisions to solve the challenges faced by any industry. For this hackathon, we have focused our efforts on an mobile app to make up for that stubborn foggy and rainy weather.

What it does

This mobile app will work in four simple steps: 1) It receives real time image frames from a camera located just in front of the train. 2) The image frames are then passed to a deep learning model which removes bad weather characteristics (e.g. mist and rain) 3) The app shows in real time how the path would be seen in normal weather conditions 4) In addition, the app will show (also in real time) a map pointing the current train location and will inidicate the next safe point (map location)

How we built it

Our deep learning model was built with PyTorch and is based on a CNN configuration with two phases, Encoder and Decoder. The first one comprehends the bluerring of normal weather images (mist, rain) which are passed to the decoder to make them back to normal.

Challenges I ran into

Improve prediction accuracy as the dataset we worked with was relatively small (2241 images).
Convert our Keras model into TensorFlow Lite (for android purposes) due to our lack of experience in machine learning for smartphones.
One member is in a different time zone.
Work with a multidisciplinary team (comp science, engineering and business)

Accomplishments that we are proud of

We all improved our teamworking and communication skills.
Our medicine student member became familiar with artificial intelligence in the health sector.
Our business student member learnt how to build android apps and got UI design experience.