Facial Recognition based Attendance System
To build a robust and stable system to record attendance Which can overcome the below drawbracks.
- Taking attendance manually with a sheet of paper by calling names during lecture hours.
- Time consuming
- Involves Manual work
- Adopted biometrics system such as fingerprint, RFID card reader to mark the attendance.
- There is a chance of card loss or an unauthorized person may misuse the card for fake attendance.
- In biometrics such as fingerprints are not recommended due to COVID-19.
A facial recognition based attendance system is a software designed to solve the issues of existing manual systems by marking the attendance through recognizing faces and automatically generating attendance reports.
- MTCNN
- pillow
- keras
- tensorflow
- sklearn
- Clone the repository
- Place the dataset folder parallel to app.py file
- Run the application with the command "flask run"
Once the application opens, a student mark his / her attendance for a particular class or an admin can login with admin credentials
An admin can do the following operations :
- Add - click on add, choose image and add an image in dataset when new student joins the class
- Delete - delete an image in dataset when a student leaves the class
- Update - click on update, choose image and update an image in dataset to update the face
On doing these operations, the train dataset gets modified and the model can recognize faces from the updated dataset
