Abstract
The project aims to develop an automatic quality assessment system for PDO dry-cured ham using 2D-3D imaging and hyperspectral analysis combined with machine learning. The goal is to classify fresh hams for visual defects and predict traits like fat thickness and weight loss during curing. It also aims to identify genes associated with meat quality through RNA sequencing and genotyping. The system will offer the ham industry a tool for monitoring the process and integrating new knowledge into genetic evaluations.
Results achieved
The project successfully established the experimental framework and collected the data required for the planned analyses. Preliminary results demonstrated the feasibility of integrating imaging technologies, hyperspectral analysis, and genomic approaches for quality assessment of dry-cured ham. Initial machine learning models showed encouraging performance in classifying quality-related traits and defects. The generated datasets and analytical pipelines provide a solid basis for further optimization and validation. Overall, the results are promising and support the development of future studies aimed at improving quality control and genetic selection in the dry-cured ham industry.Dettagli del progetto
Responsabile scientifico: Giuseppina Schiavo
Strutture Unibo coinvolte:
Dipartimento di Scienze e Tecnologie Agro-Alimentari
Coordinatore:
Università degli Studi di PADOVA(Italy)
Contributo totale Unibo: Euro (EUR) 78.130,00
Durata del progetto in mesi: 24
Data di inizio
12/10/2023
Data di fine:
28/02/2026
