Swordfish (Xiphias gladius) is an important commercial species with an extensive seasonal migration and a circumglobal distribution. In the last twenty years, catches have decreased by almost 50% and too many juveniles are caught before they reach gonadal maturity. According to the most recent stock assessment of the International Commission for the Conservation of Atlantic Tunas (ICCAT), the Mediterranean swordfish stock is overfished and suffering overfishing. In this light, comprehensive information on gonad development such as the time of spawning and the size at maturity became necessary to determine their reproductive potential in the Mediterranean Sea. The goal of this thesis is to collect and classify images of mature and immature swordfish oocytes. The main challenges are to find a method for automatic recognising the cells. In this context, the candidate will be required to study, investigate and apply machine-learning and deep learning approaches for image processing and classification.
Aim: Applying machine learning and deep learning for image classification of liver and gonal cells of swordfishes.
Supervisors: Prof. Emanuele Frontoni, Prof. Oliana Carnevali, Marina Paolanti
Start: Sept, 2018
Expected graduation: Feb/July, 2019
Skills that will be acquired: Programming (Python, MATLAB)
Contacts: {e.frontoni,m.paolanti}@univpm.it