Group Face Recognition Smart Attendance System Using Convolution Neural Network

Abstract

One of the major issues in the modern environment of complex data systems is authentication; several strategies are used to address this issue. Face recognition is regarded as one of the most dependable solutions. This research proposes a convolution neural network (CNN) for face detection and recognition that outperforms existing methods. To extract acceptable features from images, machine learning approaches necessitate expert knowledge and experience. To categorize images in an automated manner, a proposed deep learning-based strategy can be employed, which uses channel wise separable CNN to extract image features and also uses Support Vector Machine (SVM) and Softmax classifiers to classify the images. Face recognition was used to verify the accuracy of the proposed system by tracking student attendance. The public of the market tagged faces in the wild (LFW) dataset is used to train the face recognition system. On the testing data, the proposed system had a 98.11 percent accuracy rate. Furthermore, the data created by the smart classroom is processed and transferred through the use of an IoT-based edge computing approach.