作者单位:Lingraj Dora (20) Sanjay Agrawal (21) Rutuparna Panda (21)
20. Electrical & Electronics Engineering Department, VSS University of Technology, Burla, Odisha, India 21. Electronics & Telecommunication Engineering Department, VSS University of Technology, Burla, Odisha, India
ISSN:1611-3349
文摘
Regularized-LDA (R-LDA) is a LDA-based method used for finding the nonzero eigenvalues and the corresponding eigenvectors. This paper presents a RLDA-based classification of face images which uses bacteria foraging optimization (BFO) algorithm for selecting the optimal discriminating features. The optimal features are then used by probabilistic reasoning model for classification of unknown face images. The ORL and the UMIST databases are used for experiment to demonstrate the performance of our proposed method. It is observed that our proposed method outperforms the existing method.