The visual aspect of food is a problem very often neglected for different reasons; one being the difficulty of modelling vision. However, it has been shown that , on simple synthetic images, automatic sorting in the same manner as human vision is possible.
Distribution heterogeneity of spots within an image is one important parameter for its characterization. A new algorithm was tested over simple numerical pictures that gave good agreement with visual appreciation. Images of twenty dry sausage slices were analyzed and different factors corresponding to aspect structure were calculated. Principal component analyst:s based on these factors all owed discrimination between the different types of sausage. The results obtained when studying the meat part of the slices were well correlated with the visual aspect.
Roudot, A-C.; Duprat, F.; Grotte, M.-G.; and O'Lidha, G.
"Objective Measurement of the Visual Aspect of Dry Sausage Slices by Image Analysis,"
4, Article 7.
Available at: http://digitalcommons.usu.edu/foodmicrostructure/vol11/iss4/7