Contact
joseph.garrett@ntnu.no

NTNU's [HYPSO-1](https://www.ntnu.edu/web/smallsat/ntnu-smallsat-lab) satellite records images of the ocean in 120 spectral bands which can resolve many unexpected features in the earth's oceans. One of the objectives of the satellite mission is to monitor harmful algal blooms. Unfortunately, there are only small amounts of ground truth data for algal blooms. The goal of this project is to develop a machine learning algorithm that can learn to detect algal blooms from very small amounts of training data. 

This algorithm can be built upon independent component analysis, which is a group of techniques for finding the transformations of a dataset which produce the statistically unusual patterns in the data. This project would look into how small amounts of ground truth data can be incorporated into the algorithm. 

The project is related to a collaboration with Grieg Seafood and Salmar, two fish farming companies. It would be possible to join meetings with the whole collaboration and perhaps even present your work at them, if desired. It might even be possible to visit some fish farms. In addition, they have provided the ground truth data, so it will be necessary to [sign confidentiality and transfer of right agreements](https://i.ntnu.no/wiki/-/wiki/English/Student+and+business+cooperative+agreements).

Links
https://griegseafood.com