Directional Morphological Operators to Detect Fat Streaks in Iberian Ham

Iberian ham presents superior sensorial quality than other ones. Currently, chemical procedures are being used to determine fat level of the ham, but they can not give any information about its distribution. Some research made on Iberian hams from pigs with different feeding reveal similar fat level, but pigs fed on acorns and grass reached better qualifications in a sensorial analysis of the samples carried out by a trainer tester set. So, ham marbling, i.e. number, position, shape and size of intramuscular fat streaks in ham must influence its sensorial characteristics. In this work, digital image processing techniques based on mathematical morphology are proposed to achieve automated detection of fat streaks in commercial ham slices. A statistical study evaluating performance of the proposed methods is provided.