Attributes Reduction Applied to Leather Defects Classification

This paper presents a study on attributes reduction, comparing five discriminant analysis techniques: FisherFace, CLDA, DLDA, YLDA and KLDA. Attributes reduction has been applied to the problem of leather defect classification using four different classifiers: C4.5, kNN, Na\"{i}ve Bayes and Support Vector Machines. The results of several experiments on the performance of discriminant analysis applied to the problem of defect detection are reported.