Feature Extraction and Classification-Based Face Recognition Using Deep Learning Architectures

Abstract

Computer vision research is currently focused on face identification. One of the biometric systems with the quickest recent growth is this one.This paper proposed technique in face recognition using deep learning based technique in feature extraction and classification. Initially data has been processed for noise removal and image resize, then to segment the image for smoothening. Then to extract the features using Scale Invariant Feature Transforms and classified using deep belief networks. Deep learning is a way for performing facial recognition, and given its great accuracy, it appears to be a suitable technique. The proposed facial recognition system's accuracy is shown through experimental findings. The classified output shows face features and parametric analysis has been carried out in terms of accuracy, precision, recall and F-1 score for face dataset.