Name:
Spectrometric technology and methodology prototype for preventative oyster
quality assessment on-site

Developed by:
Benco Baltic d.o.o. – Croatia
UAB Art 21 – Lithuania
Spectrometric technology and methodology prototype for preventative oyster
quality assessment on-site – Lithuaina

Country(ies) involved:
Croatia

Funding sources:
EUREKA /
OWN

General idea:
The project aims to develop and demonstrate in operational environment a
technology prototype for preventative oyster quality assessment on-site, by
combining near-infrared spectrometry (NIR) with machine learning and advanced
data analysis techniques

Description:
During the project a technology prototype for the determination and estimation of protein, glycogen, fat and moisture concentrations in oysters will be developed, based on NIR spectroscopy and machine learning data analysis algorithms. The technology will include an accurate and detailed methodology for oyster sample preparation, hardware technical parameters description and specifications needed for research, analysis performance procedure, data gathering, processing and
analysis solution documentation. The machine learning algorithms and software tools (digital solutions) for precise spectral data analysis will be included in the system.
The goal of the project is to create the technology which will enable user to analyse the protein, glycogen, fat and moisture levels in oysters with the accuracy/prediction rate higher than 0.9 rate (r2) and with the prediction errors which do not exceeds 0.4 % values. The time of one analysis (including data gathering, processing and analysis) done with the prototype will not exceed 1 hour.

TRL:
7. System prototype demonstration in operational environment