Monitoring the ripening process of Iberian ham by computer vision on magnetic resonance imaging

This paper explores the use of MRI (Magnetic Resonance Imaging) in combination with a fully automated Image Analysis method for the recognition of Biceps, Femoris and Semimembranosus muscles in Iberian ham. A quantitative description of volume and a study of moisture and weight relationships during the product's ripening process are included. Three Active Contour methods (Variational Calculus, Dynamic Programming, and Greedy Algorithms) are used to recognize the Biceps Femoris and Semimembranosus muscles by means of Computer Vision techniques. The recognition of both muscles via MRI entails a low error rate (3-10%). A loss of weight in hams during the ripening process is related to a decrease in size (r(2) = 0.992). The high correlation implies that the information obtained by means of Computer Vision techniques can be used as a non-invasive complement to the traditional processes of ham weighing and moisture estimation. (c) 2007 Elsevier Ltd. All rights reserved.