Automatic attendance system based on CNN–LSTM and face recognition

In the era of the Covid-19 pandemic, the education system has undergone a significant shift from traditional offline classes to online modes of learning for students. However, keeping track of student attendance, both in offline and online modes, has become a time-consuming task. To address this issue, various automated attendance systems have been proposed, utilizing technologies such as biometric recognition, barcodes, QR codes, and mobile devices. However, earlier algorithms used in these systems have shown poor accuracy and inefficient processing times. To overcome these limitations, this paper proposes a method that leverages face recognition technology, specifically a combination of convolutional neural network (CNN) and long short-term memory (LSTM) models, to track course attendance. The combination of CNN–LSTM in attendance system is novel concept. This proposed method can be applied in both the online and offline phases of education with minor modifications. The proposed method captures both spatial and temporal information, resulting in improved accuracy. The experimental findings reported in the paper demonstrate the effectiveness of the proposed attendance system. It achieves an impressive face recognition accuracy of 99.82%, surpassing the performance of state-of-the-art methods used in similar applications with faster runtime of 6.93 s. This high accuracy rate implies that the system can reliably identify and track student attendance, ensuring accountability and providing valuable data for academic institutions. © The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management 2023.