Computer Engineering Thesis Project (15HP/30HP): Fish Monitoring using Computer Vision

Motivation and scope

The goal of this project is to develop computer vision-based techniques to assist in the monitoring and analysis of fish. These techniques could help researchers in biology and environmental chemistry by automatising work that would otherwise need to be performed manually.

Specific tasks

Your task is to develop methods and software for to assist in monitoring and tracking zebrafish within the laboratory fish tanks in the Man-Technology-Environment research centre (MTM) at Örebro University.

You will also be involved in designing a suitable camera setup for the tanks and acquiring data that can be used for the fish monitoring methods.

There are a number of different possible tasks, for example:

• Counting the number of fish in the tank
• Measuring the size of the fish in the tank
• Uniquely identifying different fish
• Tracking individual fishes within the tank
• Distinguishing male and female fish
• Evaluating anxiety or ill-health in the fish

Some of these tasks are more challenging than others and not all can be achieved within a single project. You can focus on one (or two) of these tasks.

Necessary skills

Computer engineering: Programming skills in Python or other languages. Knowledge of computer vision and/or machine learning is a benefit, but not mandatory.

Benefits

The project will encompass techniques of modern computer vision, machine learning, and object tracking which can be useful in many scenarios. The project will also contribute to ongoing research in environment health that can help both people and animals in future.

Contact

Andrey Kiselev and Stephanie Lowry

andrey.kiselev@oru.se, stephanie.lowry@oru.se