Improvement of face recognition performance using a new hybrid subspace classifier

Multiple classification systems play an important role in increasing recognition performance, especially when using heterogeneous classifiers that effectively improve performance. In this study, a new hybrid classifier was designed Fisherface and discriminative common vector approach (DCVA) subspace methods, which gave successful results face recognition. While the process of DCVA is based on properties signals belonging to classes, different signals. To create classifier, called Hybrid DCVA-Fisherface, classifiers' decision rules were combined Minimum Proportional Score Algorithm Recognition Update Algorithm. addition proposed classifiers, convolutional neural networks, Transform learning-Alexnet, Alexnet + SVM, KNN used for classification. Studies conducted ORL, YALE, Extended YALE B Face Research Lab London Set (FRLL). better examine efficiency algorithms, tests also carried out by downsampling images. When experimental analysed, higher rates than all B. However, deep learning methods generally achieved performance FRLL database, has more classes other databases.