Automated modular system for cutting and classifying frozen tuna using hyperspectral characterization (SpecTUNA)

SpecTUNA was a two-year project funded by the European Maritime and Fisheries Fund (EMFF), aimed at providing a solution to automate the timely processing of bulky frozen tuna pieces with irregular shapes, optimised cuts, and minimised discards.

Tangible results: The project has built the first 1:1 scale demo unit of SpecTUNA in basis of an existing prototype. This solution consisted of four modules: (i) the fish bin picking, consisting of a robot with artificial vision able to extract frozen tuna randomly located within a container; (ii) the cutting as usual technique (heading and tailing the fish); (iii) characterization of the food with a Hyperspectral Vision Technology System, so that organoleptic parameters (amount of salt, protein or moisture in the fish) can be determined on line in a non-invasive way; and (iv) the classification by taking advantage of the measures of these parameters.

Application/relevance: This solution has fostered the adoption of Industry 4.0 at the fish processing sector and allowed for progress in the fight against accidents at work, as well as for improvement of working conditions in the fish processing sector. It has also contributed to provide consumers with high-quality, highly nutritional, and trustworthy products by developing hyperspectral measurement of all fish canned.