Gabor feature based classification using the enhanced fisher linear discriminant model for face recognition

Area of Science:

• Computer Vision
• Biometrics
• Machine Learning

Background:

• Face recognition systems face challenges with variations in illumination and facial expressions.
• Existing methods like Eigenfaces and Fisherfaces have limitations in handling these variations effectively.

Purpose of the Study:

• To introduce a novel Gabor-Fisher classifier (GFC) for enhanced face recognition.
• To develop a robust method for face recognition that is invariant to illumination and expression changes.
• To evaluate the GFC method against other established face recognition techniques.

Main Methods:

• An augmented Gabor feature vector is derived from Gabor wavelet representations of face images.
• The enhanced Fisher linear discriminant model (EFM) is applied for dimensionality reduction, optimizing for compression and generalization.
• A Gabor-Fisher classifier is developed for multi-class face recognition problems.

Main Results:

• The novel GFC method achieved 100% accuracy in face recognition tests.
• The GFC method demonstrated high performance using a reduced feature set of only 62 features.
• Comparative studies showed the GFC method outperforming Eigenfaces, Fisherfaces, and Gabor wavelet methods.

Conclusions:

• The Gabor-Fisher classifier (GFC) presents a highly accurate and efficient solution for face recognition.
• The GFC method is robust to variations in illumination and facial expressions.
• The GFC method offers a significant advancement in face recognition technology, particularly for real-world applications.