Classification of Pre-sliced Ham Images with Quaternionic Singular Values Using an Adaptive Multilayer Perceptron

The quaternionic representation of ham images, treating RGB colour components as a single unit instead of as separate components, is very effective. The advantage of using quaternion arithmetic is that the perceptually richer colour images can be represented and analyzed as a single entity, improving the accuracy of pattern recognition models. The quaternionic singular value decomposition (SVD) is a technique to decompose a quaternion matrix into quaternion singular vector and singular value component matrices exposing useful properties. Singular values describe completely and univocally the intrinsic information of a quaternionic matrix, ergo they can be used as features for the classification of pork ham slices. The objective was to use a small portion of uncorrelated singular values, as robust features for the classification of sliced ham images, using a supervised artificial neural network classifier. Images were acquired from four qualities of sliced cooked pork ham typically consumed in Ireland (90 slices/quality), having similar appearances. Mahalanobis distances and Pearson product moment correlations were used for feature selection. The dimensionality reduction procedure excluded atypical features and discarded the redundant information. An adaptive multilayer perceptron classifier was successfully employed, using a reduced feature space of six singular values. The overall correct classification performance for the test set was 86.1%. Results confirmed that the classification performance was satisfactory. Using the most informative features as input to the multilayer perceptron classifier led to the recognition of a set of different but visually quite similar textural patterns.