Automatic Attendance System Using Face Recognition
Student attendance in the classroom has been recorded manually by the instructor so far. However, the recent developments in the field of artificial intelligence, computer vision as well as embedded systems has made such tasks to be done autonomously. This project intends to deve
2025-06-28 16:30:25 - Adil Khan
Automatic Attendance System Using Face Recognition
Project Area of Specialization Artificial IntelligenceProject SummaryStudent attendance in the classroom has been recorded manually by the instructor so far. However, the recent developments in the field of artificial intelligence, computer vision as well as embedded systems has made such tasks to be done autonomously. This project intends to develop a standalone hardware/prototype for the classroom attendance system which records the attendance of the students in an automatic way using facial recognition. The face detection algorithm will detect the faces in the image captured by the camera installed in the classroom. Next, the face recognition system will recognize the person to which the face belongs. Finally, the system will save the attendance of the present students in a microsoft excel file, to be saved at the system or share via bluetooth/email. The whole system is meant to be a standalone prototype which (after development) will be installed in a single class room for testing purposes. Such a system will save precious time for the instructor and the students. This project will provide an identity tracking solution for multi-environment applications such as official meetings, special events and casual gathering.
Project ObjectivesFollowing are the softwares/hardware based objectives for this project.
- To develop/select and optimize a suitable face detection algorithm which is robust to noise, illumination, scale and appearance.
- To count the number of faces in the image/video accurately.
- To select/develop and optimize a face recognition algorithm which is reliable, efficient and fast enough to be implemented on a single board computer such as Raspberry Pi. Such algorithm will be able to recognize multiple faces in a single image.
- To develop a prototype of the Automatic Attendance System which may be installed as a standalone portable attendance system composed of single board computer, a usb camera, and a monitor for display.
Once the prototype is developed and tested, it'll be able to be tailored for different set of conditions for human identification using face recognition.
Project Implementation MethodFacial detection and recognition is a challenging task as faces in real-world images present a very high degree of variability in scale, pose, appearance and illumination. We intend to use several algorithms for different tasks:
Face Detection Task: For face detectiuon, we'll employ and test several face detection algorithms including Viola Jones algorithm, Dual Shot Face Detector (DSFD) and Retina Face. Finally a method will be selected with higher efficiency and robustness as well as low computational cost, for reasonable real-time processing.
Face Recognition: Face recognition have been a highly active research area in last 10-15 years among computer vision society. Many techniques have been developed for face recognition incluing template-based approach, features-based approach, machine learning based approach and finally deep learning approach. We shall employ multiple approaches such as feature-based recognition, and deep learning apprach such as FaceNet. We'll focus to adopt deep learning approach since it is robust and efficient. However, final algorithm will be selected based on real-time testing efficiency and speed.
A camera will be installed in the class room.
The Raspberry Pi will take the live feed from the camera.
The live video feed captured by the camera will be processed for real-time face detection and recognition.
In addition, individual frames will be captured periodically (e.g once in 2 min), and attendance will be recorded based on face recognition applied on that captured frame.
This file will contain the attendance record for individiual student at multiple time stamps.
The attendance file may be shared with the instructor via bluetooth, or email using WiFi module.
The prototype will be tested for preferably live video processing, however, periodic frame processing will be adopted if live processing shows low efficiency with respect to accuracy and speed.
Benefits of the ProjectIt will replace the classical manual attendance system in the class.
This system will record the students' attendance seemlessly and autonomously.
It will save the precious time for the instructor as well the students.
Such a system can be tailored to be used on other venues such as to record the attendance/presense of members during formal meetings where manual attendance is inappropriate. Similarly unofficial gathering or special events can be monitoered for specific person's presence/identification.
For small/medium level companies,It'll provide an alternate and autonomous biometric attendace system other than existing finger print recognition.
Technical Details of Final DeliverableA usb camera (attached with Raspberry Pi) with adequately good resolution to get the live feed of the classroom.
A raspberry Pi single board computer with graphical interface, Buetooth and WiFi module to process the live feed captured by the camera, and save the attendnce.
An LCD monitor to preview the live feed, the processing results of the face recognition system, and the attendance record.
Final Deliverable of the Project HW/SW integrated systemCore Industry EducationOther Industries IT , Health , Security Core Technology Artificial Intelligence(AI)Other Technologies NeuroTech, Others, Big DataSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 38000 | |||
| Raspberry Pi 4, 4GB | Equipment | 1 | 17000 | 17000 |
| USB Camera | Equipment | 1 | 5000 | 5000 |
| LCD Monitor | Equipment | 1 | 11000 | 11000 |
| Report/Project assessories/Travel/Miscellaneoous Expenses | Miscellaneous | 1 | 5000 | 5000 |