Face Recognition Based on Shallow Convolutional Neural Network Classifier

Abstract

Fast and accurate user identification and verification is always desirable. Face recognition, which is machine recognization of person face by analysing patterns on facial features is becoming important for security and validation. Less interaction from user contributes high enrolment as well as easily applicable for current technology further adds its importance. In this regard, we propose a Convolutional Neural Network (CNN) based face recognition technique previously done with eigenfaces[8] but CNN has better accuracy. Entire process is divided into four phases: capturing the image, features extraction, classification and matching. At&t faces dataset is used in this paper. Input images are first fed for face detection. Face detection in input images are performed using Viola Jones algorithm. Convolutional Neural Network (CNN) is applied for feature extraction and classification. The result obtained in this paper shows that - recall is 0.992, precision is 99.4, f1 score is 99.1 and f1 beta score is 0.992 for 70-30 split of dataset, i.e, 70% dataset used as training dataset and 30% as testing dataset.