Quick menu


Summer school

Project name
Deep learning application for fish behaviour monitoring
MSc. Oleksandr Movchan
Description of the activities of the project
  • At this moment, neural networks are one of the most common methods in image and video processing. Fish behavior is correlated with the fish state and with the environmental changes in the fish surrounding. It is therefore influenced by the stress, water pollution, disease, manipulation or feeding. The application of the modern deep learning methods could solve the problem of fish behavior parametrization without the need to detect and track individual fish. This project is an attempt to classify fish school swimming patterns into several classes using deep learning.
  • The project is focused on data collection, preparation and processing to be able to distinguish between different fish behaviour.
  • The students will work with Python or Matlab for data processing and network training.
  • The outcome of the project will be trained model for behaviour monitoring.
  • Number of students - 2
  • high school students students - NO
  • university students - YES
  • Prepositions: Matlab, Image processing.

Academic and University Center
Zamek 136, Nove Hrady

Responsible person:
Webmaster: Michal Macho