A Novel Method for Attendance Marking System Using Hybrid LSTM and RNN Based Networks
E. Manigandan, Mohammed Saleh Al Ansari, Praveena Nuthakki, S Bhuvneshwari, G. Priyadharshini, Muruganantham Ponnusamy · 2023
The suggested system uses face detection and identification algorithms to automate the attendance marking and management processes. Face recognition refers to the process of identifying a person by their distinctive facial traits. At the moment, face recognition software is the most rapidly developing field in IT. This suggested system seeks to replace manual attendance tracking with an automated system that uses facial recognition technology to keep track of which students are present in the classroom at any given time. The primary goal of this endeavor is to create an automatic, user-friendly system for recording and managing attendance. The suggested method involves analyzing data, selecting features to employ, and assessing the model's efficiency. The suggested method employs Gaussian blur, segmentation, and scaling for preprocessing. PCA and LDA are used for feature selection and extraction, respectively. The LSTM-KNN hybrid method is used for model training. When compared to LSTM and KNN, two existing approaches, the proposed methodology performs exceptionally well.