Intelligent Attendance Marking System
There should be some record of attendees who attends the class or lecture in any educational institution, and here comes the use of attendance which is almost taken in every organization either manually or automated (scanning the ID card). But taking attendence manually is too hectic and consumes al
2025-06-28 16:27:56 - Adil Khan
Intelligent Attendance Marking System
Project Area of Specialization Artificial IntelligenceProject SummaryThere should be some record of attendees who attends the class or lecture in any educational institution, and here comes the use of attendance which is almost taken in every organization either manually or automated (scanning the ID card). But taking attendence manually is too hectic and consumes alot of time. Our project is being presented as a solution to this problem, moving towards our project which is to collect student's data by capturing their pictures through camera which is mounted over the blackboard, detect and recognize the faces of students by appling some techniques and algorithms and then mark their attendance automatically who are present in the class in one shot. It saves the time of both teachers as well as students students and reduce the paper work. we searched several face detection and face recognition algorithms which are suitable for our project and tried those algorithm on our actual data in order to explore them .We tried face recognition using opencv and computer vision, CNN, Haar Cascade, LPBH, which gives accurate results upto great extent. Lastly we tried GUI application which is based on PyQT. It is about how output screen will look like when program is being run,their buttons,fonts and display. For final model we decided to go with LBP, haar cascade classifier techniques which are computer vision techniques. LBP is used for face recognition and haar cascade classifier is used for face detection. The reason to go with this technique is that it gives high accuracy as compared to SVM and deep learning model (CNN). Further we decided to go with multiple face detection and extend this by training it for our collected data as it recognizes most of the student's faces in a single picture and give good results.
Project Objectives- To make the attendance marking and management system fully automatic, efficient, time saving, simple and easy.
- To reduce the workload of people and also saves times.
- To develop an automated system that marks the student's attendance by using facial recognition technology.
- To overcome that difficulty if any student is wearing mask or glasses, the trained model would also mark his/her attendance correctly.
One camera is mounted simply over the blackboard which will detect and catch the images of the students of whole class in one single picture. It will compare the image of every student with the pre-stored images in the database and if the student's face is matched with the already predefined images in the database the student will be marked as present and the remaining ones that are not present in the class will be marked as absent. The processed image will then be compared against the existing stored record and then attendance is marked in the database accordingly. Compared to existing system traditional attendance marking system, this system reduces the workload of the people .This proposed system will be implemented in 4 phases such as Image Capturing , Segmentation of group image and face detection , Face comparison and recognition , Updating of attendance in database.
- Data Collection: We will collect data by using camera which will be set in the classroom.
- Tools Exploration: Tools will also be selected on the basis of programming language that will be used. Most probably, Google COLAB and VISUAL STUDIO CODE will be preferred because their interfaces are quite descriptive that may help in demonstration. After the completion of model wordpress could be used for the tools and libraries to develop a web-server platform.
- Modeling and Training: An efficient model will be chosen after testing different models/classifiers like Linear Regression, Random Forest and some other ensemble models like XGBoost etc. As predictive analysis is totally based on the accuracy of model that will be predicting the values, these given training model are being used now a days and then the most accurate one is selected after comparative analysis.
- Comparing Results and Performance Analysis: A comparison will be done on the basis of visualization and graphs after getting results from different classifiers and then the best classifier will be chosen for the prediction model.
- Deploying:The proposed model will might be deploying our model on server with a front-end.
- To help the lecturers, improve and organize the process of tracking and managing student attendance
- Provides a valuable attentive service for both teachers and students
- Produce monthly reports for lecturer
- Reduce time loss as time is very valuable resource
- This system also motivate students and improve the academic performance of the students
- Increased accuracy and reliability
- Increased operational efficiency
- Data security
- Easy to manage
The final deliverable records attendance of whole class in one shot by using face detection and face recognition techniques and for this purpose a camera is used in the classroom that takes images of students and marked the attendance of the whole class correctly which are present in the classroom.
For final model we decided to go with LBP, haar cascade classifier techniques which are computer vision techniques. LBP is used for face recognition and haar cascade classifier is used for face detection. The reason to go with this technique is that it gives high accuracy as compared to SVM and deep learning model (CNN). Further we decided to go with multiple face detection and extend this by training it for our collected data as it recognizes most of the student's faces in a single picture and give good results. All algorithms which are being used in the project is in python language and in the end a front end application will be made which shows attendance marking of students and their records.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 80000 | |||
| Camera | Equipment | 1 | 12000 | 12000 |
| Printing Cost | Miscellaneous | 16 | 300 | 4800 |
| Travelling Expense | Miscellaneous | 8 | 250 | 2000 |
| Report File | Miscellaneous | 1 | 700 | 700 |
| Stationary | Miscellaneous | 5 | 500 | 2500 |
| Web Hosting Service | Equipment | 1 | 5500 | 5500 |
| Paid Courses for this FYP | Equipment | 5 | 1500 | 7500 |
| Graphics Card | Equipment | 1 | 45000 | 45000 |