The SICAPTOR 2.0 project intends to redesign the iObserver electronic monitoring (REM) system in order to minimize its size and maximize its performance. This device takes pictures of all fishes passing through the conveyor belt in the fishing park, and identifies in real time the species to which each of the specimens belongs in each of the photographs, making also an estimate of height and weight of each individual and of the total amount of captured biomass. To attain these objectives, we use a set of deep learning algorithms specifically developed.
The aim of SICAPTOR 2.0 is to increase its real installation potentialities on board fishing vessels of all types of fleets (not only trawlers) as a reference REM device. With this objective, proposed actions in the project are focused on: (i) the separation of the vision hardware (camera) from the processing device (computer), which will be integrated as one more instrument on the skipper’s bridge; (ii) likewise, and given the great advances that have been made in recent years in the field of artificial vision, it is intended to analyze and evaluate other more compact vision systems and/or with lower light requirements, such as such linear cameras, facing efficiently the replacement of current matrix cameras used and; iii) re-training of the deep learning algorithms (developed in SICAPTOR to estimate in real time the composition and volume of the total catches) towards its adaptation to the new image formats generated by the linear cameras tested and, if needed, the development of new specific ones.
