2022/04/26 by G. Daumas, Antoine Vautier, Daumas, Gérard +15
Engineering · #Metallurgy and Material Forming
paper · doi:10.17180/ciag-2022-vol85-art14
The sorting of pork cuts is based on their tissue composition and meat quality. But it is mostly manual and of limited accuracy. A magnetic induction scanner was tested in the HYPER-SCAN project on 80 bellies and 100 hams. 88% of the bellies were well classified in the 4 commercial classes. The muscle content of ham was predicted with an R² of 0.93. X-ray tomography showed that 3 or 4 measurement sites were sufficient to accurately predict the composition of hams, bellies and loins (R² > 0.90). A prototype hyperspectral imager was developed to qualify the meat quality of the loin. The ultimate pH was the best predicted criterion (R² = 0.65). The predictions of exudate, intramuscular lipid content and technological yield were found to be satisfactory. This system showed great potential for the development of a more accurate online tool for automatic meat quality prediction.