Detecting and Recognizing the Face and Iris Features from a Video Sequence using DBPNN and Adaptive Hamming Distance
Dense feature extraction is becoming increasingly popular in face recognition. Face recognition is a vital component for authorization and security. In earlier days, CCA (Canonical Correlation Analysis) and SIFT (Scale Invariant Feature Transforms) was used for face recognition. Since multi scale extraction is not possible with these existing methods, a new approach to dense feature extraction is developed in this project. The proposed method combines dense feature extraction and decision based propagation neural network (DBPNN). Neural network algorithm is presented to recognize the face at different angle, and it is used for training and learning and leading to efficient and robust face recognition. Finally Iris matching is done by using Iterative randomized Hough transform for detecting the pupil region with number of iteration counts. Experimental results show that the proposed method is providing effective recognition rate with accuracy in comparing with existing methods.
