A Study on Detection and Recognition of Facial Area Using Linear Discriminant Analysis

Abstract

We propose a more stable robust recognition algorithm which detects faces reliably even in cases where there are changes in lighting and angle of view, as well it satisfies efficiency in calculation and detection performance. We propose detects the face area alone after normalization through pre-processing and obtains a feature vector using (PCA). The feature vector is applied to LDA and using Euclidean distance of intra-class variance and inter class variance in the 2nd dimension, the final analysis and matching is performed. Experimental results show that the proposed method has a wider distribution when the input image is rotated $45^{\circ}$ left / right. We can improve the recognition rate by applying this feature value to a single algorithm and complex algorithm, and it is possible to recognize in real time because it does not require much calculation amount due to dimensional reduction.

Figure 1. Algorithm of the Whole System

Figure 2. Facial images in consideration of the directionality

Figure 3. Average image of face

Figure 4. Face distribution of eigenspace

Figure 5. Within-class and between class scatter

Figure 6. Dimensionality reduction

Table 1. Recognition compared to the existing algorithms(%)

Table 2. Comparison of recognition in the background change(%)

Table 3. Face detection by comparison to lighting changes(%)