Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 21

Smart Attendance System

Bachelor of computer application

Supervisor: Mr. MOHAMMED JALALUDDIN


(Assistant professor)

KEERTHIVASAN A DEEPAK.B
211421601014 211421601006

DEPARTMENT OF COMPUTER APPLICATIONS


PROJECT PROPOSEL

 ABSTRACT
 OBJECTIVE OF THE PROJECT
 LITERATURE SURVEY
 EXISTING SYSTEM
 PROPOSED SYSTEM
 METHODOLOGY
 ARCHITECTURE DIAGRAM
 DETAILED DIAGRAM
 IMPLEMENTATION
 TESTING PHASE
 RESULT WITH ANALYSIS
 CONCLUSION AND FUTURE SCOPE
 REFERENCES

DEPARTMENT OF COMPUTER APPLICATIONS 2


ABSTRACT

The conventional methods of taking attendance in educational institutions are


often time-consuming and susceptible to inaccuracies
The process has several shortcomings, such as requiring a long time to make
attendance; the attendance paper is lost, the administration must enter
attendance data one by one into the computer.
To overcome this, this method proposed a web-based student attendance
system that uses face recognition.
In the proposed system, Convolutional Neural Network (CNN) is used to
detect faces in images, deep metric learning is used to produce facial
embedding, and K-NN is used to classify student’s faces.
 Thus, the computer can recognize faces. From the experiments conducted,
the system was able to recognize the faces of students who did attend and
their attendance data was automatically saved.
DEPARTMENT OF COMPUTER APPLICATIONS 3
OBJECTIVE
• Accuracy and Efficiency:
• Develop a system that accurately recognizes and verifies individuals based on facial
features.
• Ensure real-time processing to minimize delays in attendance tracking.

• Enhanced Security:
• Implement a secure system to prevent unauthorized access or attempts to
manipulate attendance data.
• Utilize advanced face recognition algorithms to enhance system security.

• Real-time Monitoring and Reporting:


• Enable real-time monitoring of attendance data for administrators.
• Generate detailed reports and analytics to track attendance trends and patterns.

DEPARTMENT OF COMPUTER APPLICATIONS 4


LITERATURE SURVEY
S. AUTHOR TITLE METHODOLOGY
NO

1. Hao Yang, Face recognition attendance 1. Real-Time Video Face Image


Xiaofeng Han system based on real time Recognition
video processing 2) Face Feature Extraction by LDA
(vol 8 , pp.159143-159150 , Method
3) Main Face Recognition Methods
2020) IEEE Access (journal)

2. Nianfeng Li, Face Dataset for face 1. Face dataset obtained by camera
Xiangfeng Shen, recognition in an 2. Processed the dataset by CNN
Liyan Sun, uncontrolled classroom 3. K NN is used to classify the face in
Zhiguao environment the dataset
(vol 11 pp.86963-
86976,2023) (journal)

DEPARTMENT OF COMPUTER APPLICATIONS 5


LITERATURE SURVEY
S. AUTHOR TITLE METHODOLOGY
NO

Deep Unified Model For Face


3. Muhammad Zeeshan
Khan, Recognition Based on
1. Collecting face data
2. Validating face data using CNN
Saad Harous, Convolution Neural Network 3. Uploading final result in
Saleet Hassan and Edge Computing(2019) dashboard
IEEE Access (journal)

4. Zhigang Gao, A Student Attendance 1. Subregion Selection


Yucai Huang, Management Method 2. Crowdsensing
Leilei Zheng, Based on Crowdsensing in 3. Verification
Xaiodong Li Classroom Environment
(2020) IEEE Access (journal)

DEPARTMENT OF COMPUTER APPLICATIONS 6


LITERATURE SURVEY
S. AUTHOR TITLE METHODOLOGY
NO

5. Mrs.Aishwarya, Attendance monitoring Monitoring attendance and granting


Supriya , and access control access to authorized users
Sushmitha , system(2021) (journal)
Vandana,
Mrs. Bhargavi

6. Aparna Trivedi, Face Recognition Based 1. Collecting sample face data


Chandan Mani Tripathi, Automated Attendance 2. Training the ML model with
Dr. Yusuf Perwej Management sample face data
System(2019) 3. Validation of the face with the
(conference)
sample data
4. Final attendance uploaded in the
dashboard

DEPARTMENT OF COMPUTER APPLICATIONS 7


EXISTING SYSTEM
• Existing systems for attendance taking vary depending on the context and the specific needs
of the organization or institution.

• Some common methods include Manual Attendance Registers: This traditional method
involves using paper-based attendance registers where individuals physically sign in or mark
their attendance.

• While simple and low-cost, manual attendance registers are prone to errors, such as
illegible handwriting or proxy attendance, and can be time-consuming to manage.

• Cards: RFID cards contain embedded chips that can be scanned by RFID readers to track
attendance.

• Individuals carry RFID cards with them and swipe them at designated checkpoints to record
their presence.

• While RFID technology offers greater convenience and speed compared to manual methods,
it requires the distribution and management of physical cards

DEPARTMENT OF COMPUTER APPLICATIONS 8


PROPOSED SYSTEM

• The traditional method of taking attendance is has various shortcomings and it is time consuming and the
attendance can be faked and the attendance data should be updated manually by the teachers.

• In our proposed system we will take sample images of the person.

• Then it will be stored in local storage systems and when a person appears on a camera the face data will be
extracted and it will be detected and recognized by using KNN and CNN .

• once the face matched the attendance will be marked and it will be displayed in the dashboard .

DEPARTMENT OF COMPUTER APPLICATIONS 9


METHODOLOGY
• Data Collection and Preprocessing:
Gather a dataset of facial images representing individuals expected to be recognized by the
system.Preprocess the dataset by standardizing image sizes, adjusting brightness/contrast, and removing noise
to enhance recognition accuracy
• Face Detection:
Implement a face detection algorithm to locate and extract faces from images or video streams. Choose an
efficient and accurate face detection technique such as Haar cascades, HOG, or deep learning-based approaches
like CNNs.

• Feature Extraction:
Extract discriminative features from the detected faces to represent their unique characteristics.
Utilize feature like OpenCv , Dlib , numpy , pandas

• Face Recognition Model:


Select a suitable face recognition algorithm or model based on the requirements and available
resources.
Train the face recognition model using the preprocessed dataset and extracted features.Consider techniques such as
deep learning-based approaches like KNN and CNN

• Real-Time Processing:
Implement the face recognition model to process live video streams or camera feed in real-time.
Optimize the processing pipeline for efficiency and low latency to meet real-time constraints.
Utilize hardware acceleration techniques (e.g., GPU computing) to improve processing speed if necessary.

DEPARTMENT OF COMPUTER APPLICATIONS 10


METHODOLOGY

• Attendance Logging:
Integrate the face recognition system with an attendance logging
mechanism to record attendance data. Upon recognizing a face, log the attendance of
the corresponding individual along with relevant metadata such as timestamp and
location.

• Security and Privacy Considerations:


Implement security measures to protect sensitive attendance data and
prevent unauthorized access to the system.Address privacy concerns by adopting data
anonymization and encryption techniques as necessary

Maintenance and Updates:


Regularly maintain and update the system to address bugs, security
vulnerabilities, and evolving user requirements.Monitor system performance and user
feedback to identify areas for improvement and optimization.

DEPARTMENT OF COMPUTER APPLICATIONS 11


ARCHITECTURE DIAGRAM

DEPARTMENT OF COMPUTER APPLICATIONS 12


IMPLEMENTATION

DEPARTMENT OF COMPUTER APPLICATIONS 13


IMPLEMENTATION

Activating-the-camera- -encodings-generated-from-the-Face-recognition-library

DEPARTMENT OF COMPUTER APPLICATIONS 14


IMPLEMENTATION

DETECTING THE IMAGE AND UPLOADING IT TO A DATABASE FOR MONITORING IT

DEPARTMENT OF COMPUTER APPLICATIONS 15


IMPLEMENTATION

Identifying the person and making attendance in local CSV file

DEPARTMENT OF COMPUTER APPLICATIONS 16


RESULT WITH ANALYSIS

The final result shows the attendance dashboard , where it shows the attendance records of the students

DEPARTMENT OF COMPUTER APPLICATIONS 17


CONCLUSION AND FUTURE SCOPE

• Implementing a real-time smart attendance system using face recognition technique offers a
range of benefits and opportunities for various industries and institutions

• The automation of attendance tracking through face recognition eliminates the need for manual
processes, saving time and reducing administrative overhead. Real-time processing ensures that
attendance records are up-to-date and accurate.

• By verifying individuals based on their unique facial features, face recognition enhances security
by reducing the risk of proxy attendance or identity fraud. Access controls can be implemented to
restrict entry only to authorized individuals

DEPARTMENT OF COMPUTER APPLICATIONS 18


REFERENCES

1 Authors:M. Singhal and G. Ahmad,


Title:"Deep Learning Based Real Time Face Recognition For University Attendance System,"
Published year:2023, Japan, 2023, pp. 01-04, doi: 10.1109/ISDCS58735.2023.10153549.

2 Authors:H. Yang and X. Han,


Title: "Face Recognition Attendance System Based on Real-Time Video Processing," in IEEE Access, vol. 8, pp.
159143-159150,
Published year: 2020, doi: 10.1109/ACCESS.2020.3007205

3 Authors: B. Kocacinar, B. Tas, F. P. Akbulut, C. Catal and D. Mishra,


Title: "A Real-Time CNN-Based Lightweight Mobile Masked Face Recognition System," in IEEE Access, vol. 10, pp.
63496-63507,
Published year: 2022, doi: 10.1109/ACCESS.2022.3182055.

4. Authors: C. Lu and X. Tang,


Title: ‘‘Surpassing human-level face verification performanceon LFW with gaussianface,’’,
Published year: 2015, IEEE Access

5. . Authors: L. Wolf, T. Hassner, and I. Maoz,


Title: ‘‘Face recognition in unconstrained videos with matched background similarity
Published year: 2019

DEPARTMENT OF COMPUTER APPLICATIONS 19


CONFERENCE

DEPARTMENT OF COMPUTER APPLICATIONS 20


DEPARTMENT OF COMPUTER APPLICATIONS 21

You might also like