A Novel Hybrid Feature Extraction Technique For Face Recognition
P Naveen Kumar, B Naresh Kumar, Ginjupalli Srinivasa Rao · 2013
Face recognition is a computer for identifying and retrieving desired images from a large collection on the basis of features(color, texture, shape..).CBIR system is generally used in security, medicine, entertainment…etc. Interesting method for dimensionality reduction is subspace analysis. Images are portioned into sub images by using sub band decomposition or orthogonal transform to detect local features. Shape feature of an image can be represented using various moments. Image representation using moments has desirable properties of rotation invariance, robust to noise, expression efficiency, effective ness, and multi level representation for describing various shapes of patterns. Further in this work different distance measure like Minkowski distance, Manhatten distance, Euclidean distance etc. will be used to test the performance of the proposed methods