Abstract
Quality of the egg is examined before the
production process in the food industry. In this
study, a 3D laser scanning approach was developed to
measure internal quality traits of the
egg. Experiments were carried out on 450 chicken egg
samples in different freshness stages. Spectroscopic
analysis showed 600–700 nm would be a suitable
wavelength region for laser scanning. The
quasi-liquid transparency nature of the albumin
introduced issues to laser scanning, which were
fixed by two solutions: 1. coating of the thick
albumin region with Talcum powder before scanning,
and 2. training a multilayer perceptron neural
network to map the 2D shape features of the thick
albumin region to the actual albumin height
values. Results indicated that the yolk index and
Haugh unit are reliably measured using 3D scanning
of the yolk part (R = 0.97 and RMSE = 0.04) and the
2D shape analysis of the thick albumin region (R =
0.84 and RMSE = 6.53), respectively.
Users
Please
log in to take part in the discussion (add own reviews or comments).