Impact of Kernel Fisher Analysis Method on Face Recognition
Amruta S. Moon1, Rajiv Srivastava Yogdhar Pandey2
1Amruta S. Moony, CSE Dept, SIRT. Bhopal SIRT. Bhopal, India.
2Rajiv Srivastava Yogdhar  Pande,  Director, SI.T. Bhopal India.
Manuscript received on January 21, 2013. | Revised Manuscript received on February 02, 2013. | Manuscript published on February 28, 2013. | PP: 545-551 | Volume-2 Issue-3, February 2013.  | Retrieval Number: C1191022313/2013©BEIESP

Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: Human Face recognition is a challenging task in computer vision and pattern recognition. Face recognition is difficult because it is a real world problem. The Human Face is complex, natural object that tends not to have easily identified edges and features. Because of this, it is difficult to develop a mathematical model of that face that can be used as prior knowledge when analyzing a particular image. This paper deals with the correspondence presents Color and Frequency Features based face recognition. The CFF method, which applies an Enhanced Fisher Model (EFM), extracts the complementary frequency features in a new hybrid color space for improving face recognition performance. The new color space, the RIQ color space, which combines the component image of the RGB color space and the chromatic components and of the YIQ color space, displays prominent capability for improving face recognition performance due to the complementary characteristics of its component images. The EFM then extracts the complementary features from the real part, the imaginary part, and the magnitude of the image in the frequency domain. The complementary features are then fused by means of concatenation at the feature level to derive similarity scores for classification. The complementary feature extraction and feature level fusion procedure applies to the I and Q component images as well. The hybrid color space improves face recognition performance significantly, and the complementary color and frequency features further improve face recognition performance. In CFF method particular, the Indian database had used for experimental analysis. There are many problems with face recognition such as facial expression, pose, age and occlusion. The Training set contains 200 images that are either controlled or uncontrolled. The Target set has 400 controlled images and the Query set has 100 uncontrolled images. While the faces in the controlled images have good image resolution, the faces in the uncontrolled images have lower image resolution and . These uncontrolled factors pose grand challenges to the face recognition performance. The face images used in our experiments are normalized to 64×64 to extract the facial region, which contains only face and the performance of face recognition is thus not affected by the factors not related to face, such as hair styles. These experimental results show that the combination of the hybrid color and frequency features by the CFF method is able to further improve face recognition performance. In particular the CFF method achieves the face verification rate (corresponding to the TestSet3) of 80.3% at the false accept rate of 0.1%. Future research will be considerd applying kernel methods, such as the multiclass Kernel Fisher Analysis (KFA) method to replace the EFM method for improving face recognition performance. And Note that the KFA method achieves, at 0.1% false accept rate, 84% face verification rate (FVR) respectively. Experimental result shows that the proposed method is efficient and improves the face recognition performance by large margin.
Keywords: KFA (Kernel Fisher Analysis), CFF(Color and Frequency Features based face recognition),EFM, RIQ.