Face Recognition Using Combined Global Local Preserving Projections and Compared With Various Methods

In appearance-based methods, we usually represent an image of size n x m pixels by a vector in an n x m dimensional space. How ever, these n x m-dimensional spaces are too large to allow robust and fast face recognition. A common w ay to attempt to resolve this problem is to use dimensionality reduction techniques. The most prominent existing techniques for this purpose are Principal Component Analysis (PCA) and Locality Preserving Projections (LPP). We propose a new combined approach for face recognition which aims to integrate the advantages of the global feature extraction technique like PCA and the local feature extraction technique LPP .It has been introduced here (CGLPPCombined Global Local Preserving Projections. Finally, Comparison is done w ith various recognition methods. Experimental evaluations are performed on the ORL and UMIST data sets w ith 400 images and 40 subjects.