Research Article
Automatic Attendance Monitoring System Using Deep Learning
@INPROCEEDINGS{10.4108/eai.7-12-2021.2314601, author={Shanmuhappriya M and Sudha Sadhasivam G}, title={Automatic Attendance Monitoring System Using Deep Learning}, proceedings={Proceedings of the First International Conference on Combinatorial and Optimization, ICCAP 2021, December 7-8 2021, Chennai, India}, publisher={EAI}, proceedings_a={ICCAP}, year={2021}, month={12}, keywords={automatic attendance face recognition facenet deep learning jetson nano}, doi={10.4108/eai.7-12-2021.2314601} }- Shanmuhappriya M
Sudha Sadhasivam G
Year: 2021
Automatic Attendance Monitoring System Using Deep Learning
ICCAP
EAI
DOI: 10.4108/eai.7-12-2021.2314601
Abstract
The manual method of attendance management is time consuming and difficult to maintain. So, the process of management of attendance in schools and universities should be automated. Many biometric systems can be used to record attendance. Face recognition is used frequently. The proposed work aims to record attendance without human interference. In this method the camera is fixed in the classroom.Faces are detected from the captured image and then recognized using trained model. The attendance is then marked and verified by class teacher. The attendanceis marked and absenteeism is sent to the parents. The proposed tooluses Deep Convolution Neural Network(DCNN) in which Max-Margin Object Detection(MMOD) and Histogram of Orientation Gradient(HOG) in Dlib used for face detection. Facenet algorithm responsible for extracting high quality features from the given faces is used for face recognition. The model is trained using K-Nearest Neighbour(KNN) and the framework is loaded in Jetson nano for testing.