Improved Face Recognition Rate Based on PCA and LDA Using Image Preprocessing Techniques

Abstract

Principal component analysis (PCA) and linear discriminant analysis (LDA) are two powerful tools utilized for dimensionality reduction and feature extraction in most of pattern recognition applications. PCA selects features important for class representation while LDA algorithm selects features that are most effective for class separability. This paper carries out a comparative study of the two important algorithms in the context of enhancement obtained using certain image preprocessing techniques. The implementation has been done in the MATLAB programming environment, and its performance is investigated using facial images from ORL and created databases. The images are preprocessed to yield normalized images which are subsequently used for recognition task.

Keywords: principal component analysis (PCA), linear discriminant analysis (LDA), eigen face.

Cite this Article

Joshi Jagdish Chandra, Gupta KK. Improved face recognition rate based on PCA and LDA using image preprocessing techniques. Journal of Instrumentation Technology and Innovation. 2015; 5(2): 19–24p.