Within this work package, the participating partners, coordinated by CNR, will cooperate towards the development of an automated tool for Ostreopsis ovata identification and counting.
This goal will be reached following three steps. First of all, an opto-electronics platform for single algae imaging in 2 and 3 dimensions (called QPM) will be set up.
The partner OnAir will implement and optimize the QPM capabilities to examine, identifiy and reconstruct images.
Since the targeted algae Ostreopsis Ovata can assume a wide range of shapes, the partner CoNISMa will support the learning procedure of QPM by providing a full cell measurement dataset.
This sophisticated training will allow the tool to obtain a correct taxonomic identification based both on morphological characters and on advanced genetic signatures.
Partners will take advantage of experience in plankton identification methods, based on standard image processing approaches, in order to maximize the precision with which the tool will learn to recognize Ostreopsis.