Face Recognition System using Self Organizing Feature Map and Appearance Based Approach
Face Recognition has develop one of the most effective presentations of image analysis. This area of research is important not only for the applications in human computer interaction, biometric and security but also in other pattern classification problem. To improve face recognition in this system, two methods are used PCA Principal component analysis and SOM Self organizing feature Map .PCA is a subspace projection method is used compress the input face image. SOM method is used to classify DCT based feature vectors into groups to identify if the subject in the input image is “present” or “not present” in the image database. The aim of this system is that input image has to compare with stored images in the database using PCA and SOM method. An image database of 100 face images is evaluated containing 10 subjects and each subject having 10 images with different facial expression. This system is evaluated by measuring the accuracy of recognition rate. This system has been implemented by MATLAB programming.
PCA Principal component analysis , SOM Self organizing feature Map , DCT discrete cosine transform
Thaung Yin | Khin Moh Moh Than | Win Tun