Face recognition using principal component analysis and self organizing maps

Face recognition is a vital part of object recognition research which the scientific community has shown a growing attention in the past few decades. Since then, the rapid development of technology and the commercialization of technological achievements, face detection became more popular. One of the challenges in face recognition systems is to recognize faces around different poses and illuminations. The face recognition phases include image preprocessing, feature extraction, and clustering. This research focus on developing a face recognition system based on Principal Component Analysis (PCA) and Self-Organizing Maps (SOM) unsupervised learning algorithm. The preprocessing steps contain grey scaling, cropping and binarization. The selected dataset for this research is Essex database that are collect at University of Essex which consist of 7900 face images taken from 395 individuals (male and female). Face recognition is a vital part of object recognition research which the scientific community has shown a growing attention in the past few decades. Since then, the rapid development of technology and the commercialization of technological achievements, face detection became more popular. One of the challenges in face recognition systems is to recognize faces around different poses and illuminations. The face recognition phases include image preprocessing, feature extraction, and clustering. This research focus on developing a face recognition system based on Principal Component Analysis (PCA) and Self-Organizing Maps (SOM) unsupervised learning algorithm. The preprocessing steps contain grey scaling, cropping and binarization. The selected dataset for this research is Essex database that are collect at University of Essex which consist of 7900 face images taken from 395 individuals (male and female).