Automated Attendance Marker Using Facial Recognition

Abstract

Attendance management is a crucial task in various domains, such as educational institutions, workplaces, and events, to accurately record and monitor the presence of individuals. Traditional attendance marking methods involving manual processes are time-consuming, prone to errors, and lack efficiency. This abstract proposes a novel solution, the \"Attendance Marker Using Facial Recognition,\" which leverages the advancements in facial recognition technology to automate attendance tracking The attendance marker system offers numerous advantages over traditional methods. Firstly, it eliminates the need for manual data entry, reducing human error and increasing efficiency. It also ensures reliable identification, as facial features are difficult to forge or replicate. Moreover, the system can handle a large number of attendees simultaneously, making it suitable for events with high footfall

Introduction

I. INTRODUCTION

 Maintenance of an attendance system is crucial for monitoring student performance across all institutes in this system. Teachers use attendance sheets in the majority of the institution's traditional attendance marking system. Students will either sign the attendance form and file it or log on to the computer for later review. This method is known as tiresome. As some of the students frequently sign for their pals, this is time-consuming and inaccurate. Keeping track of every student's attendance in a big classroom is challenging. It is tiresome to keep track of students' attendance in class. The system for recording attendance includes facial recognition, stream image processing, and storage of the data in a database that is maintained by the teachers. creates a database of the faculty, staff, and students. An automated attendance marker is a system that is used to track the attendance of individuals in an organization, such as a school or business. It typically works by using some form of identification, such as a face, fingerprint, or RFID card, to verify the identity of the individual and mark their attendance in a database or on a physical attendance sheet. Automated attendance markers can provide a convenient and efficient way to track attendance, as they can save time and reduce the risk of errors compared to manual attendance tracking methods. They may also be used for a variety of purposes, such as calculating pay for hourly employees or tracking attendance for course credit.

Types: There are several types of automated attendance markers, which use different methods to identify individuals and mark their attendance. Some common types include:

Face recognition systems, which use cameras and computer vision algorithms to identify individuals based on their facial features.Fingerprint scanners, which use fingerprint scanners to identify individuals based on the unique patterns of their fingerprints.RFID (Radio-Frequency Identification) systems, which use RFID cards or tags to identify individuals when they are within range of a reader.

Advantages: Automated attendance markers can provide a number of advantages over manual attendance tracking methods. For example:They can save time and effort, as individuals do not need to sign in or out manually.They can reduce the risk of errors, such as misinterpreting handwritten names or accidentally marking the wrong person as present.They can provide more accurate and up-to-date attendance records, which can be useful for a variety of purposes (e.g., tracking attendance for course credit, calculating pay for hourly employees). They can be more convenient for individuals, who do not need to remember to bring an attendance sheet or sign in manually

II. RELATED WORKS

A. Paper 1: Recognition based Attendance System for Classroom Environment

1. Enrollment: The first step is to enroll the students in the system. This typically involves capturing images of their faces and storing them in a database along with their identifying information (e.g., name, ID number).
2. Attendance marking: To mark attendance, the system captures an image of each student's face as they enter the classroom using a camera.

The system then extracts features from the images using a facial recognition algorithm and compares them to the enrolled students' images in the database. If a match is found, the system can mark the student's attendance in the database or on a physical attendance sheet.

3. Accuracy: The accuracy of the attendance marker will depend on a variety of factors, including the quality of the images captured, the size of the enrolled population, and the similarity of the enrolled students. In general, systems that use facial recognition tend to be more accurate when the enrolled population is small and the students are relatively distinct from one another.

Advantages

• It can save time and effort compared to manual attendance tracking methods.
• It can reduce the risk of errors, such as misinterpreting handwritten names or accidentally marking the wrong student as present.
• It can be more convenient for students, who do not need to remember to bring an attendance sheet or sign in manually.
• It can provide more accurate and up-to-date attendance records, which can be useful for a variety of purposes.
• Disadvantages
• Cost: Implementing the system may be expensive.
• Privacy concerns: Some people may have concerns about the collection and processing of personal biometric data.
• Accuracy: The system may not always be 100% accurate.
• Dependence on technology: The system may be prone to technical issues or failures.
• Limited to classroom use: The system is typically only practical for use in a classroom setting

B. Paper 2: Automatic Attendance System Using Face Recognition By Viola Jones Algorithm

1. Pre-processing: The input image is first converted to grayscale and resized to a smaller resolution to reduce the computation time.
2. Feature extraction: The algorithm extracts Haar-like features from the image. These features are simple pixel patterns that can be used to distinguish between faces and non-faces.
3. Feature selection: The algorithm selects the most important features from the previous step using Adaboost, a machine learning algorithm that selects a subset of the features that are most effective at classifying faces.
4. Training: The algorithm trains a classifier using the selected features and a set of labeled training data (i.e., images of faces and non-faces).
5. Detection: To detect faces in a new image, the trained classifier is used to compute the scores for each Haar-like feature in the image. The algorithm then combines the detections from multiple scales and merges overlapping detections to give the final set of face detections.

         Advantages

• Fast: The algorithm is relatively fast and can be used in real-time applications.
• Robust: It is robust to changes in lighting, pose, and facial expressions.
• Efficient: It is efficient in terms of both memory usage and computational time.
• Widely used: It has been widely used in many applications and has been extensively tested and refined over the years.

• Open source: The algorithm has been released as open source software, making it freely available to researchers and developers.
• Disadvantages
• Limited to frontal faces: The algorithm is limited to detecting frontal faces and may not work well for profiles or faces at other angles.
• Sensitive to scale: The algorithm is sensitive to the scale of the face in the image, so the face must be roughly the same size as the training data.
• Poor performance on some face types: The algorithm may perform poorly on certain types of faces, such as those with unusual features or facial hair.
• Complexity: The algorithm is somewhat complex, with multiple stages and parametersthat must be set properly in order to achieve good performance.
• Limited to two-class problem: The algorithm is limited to a two-class problem (face or non-face) and cannot be used to identify specific individual.

C. Paper 3: Class Room Attendance System Using A3D Facial Model

a. Methodology

1. Set up a camera at the front of the classroom that is capable of capturing 3D facial images
2. Use software or a library that is able to detect and recognize faces in the 3D images captured by the camera. There are a number of open source options available for this, such as OpenCV or dlib.
3. Create a database of 3D facial images for each student in the class. This can be done by having each student pose for the camera and capturing their 3D facial image.
4. Train a machine learning model to recognize the 3D facial images of each student in the database. You can use a supervised learning approach for this, where you provide the model with a labeled training dataset consisting of 3D facial images and the corresponding student labels.
5. Set up the camera and facial recognition software to automatically capture 3D facial images of students as they enter the classroom.
6. Use the trained machine learning model to recognize the students in the captured 3D facial images and mark them as present in the attendance record.
7. To ensure accuracy, you may want to set up a system for manual verification of attendance. For example, you could have a teacher or TA verify the attendance record and make any necessary corrections.

b. Advantages

• Improved accuracy: 3D facial recognition can be more accurate than 2D facial recognition, particularly in challenging lighting conditions or when the face is partially occluded. This is because 3D facial recognition systems capture the full depth and contour of the face, which allows them to identify more distinctive features.
• Robustness to pose and expression: 3D facial recognition systems are less sensitive to changes in pose or expression than 2D systems, which can make them more reliable in real-world applications.
• Enhanced security: Because 3D facial recognition systems capture more detailed and accurate information about the face, they can potentially offer better security than 2D systems. For example, a 3D facial recognition system might be more resistant to spoofing attacks, in which an attacker tries to gain access to a system by presenting a photo or video of a person's face.
• Improved usability: In some cases, 3D facial recognition systems may be easier for users to interact with than 2D systems. For example, a 3D system might be able to recognize a person's face from a wider range of angles, which could make it easier for the user to use the system.
• New applications: The additional data captured by 3D facial recognition systems may enable new applications that are not possible with 2D systems. For example, 3D facial recognition might be used to create more realistic avatars or to automatically generate 3D models of people's faces.

c. Disadvantages

• Cost: 3D facial recognition systems can be more expensive to develop and deploy than 2D systems, both in terms of hardware and software. This can be a significant barrier to adoption, particularly for small businesses or organizations with limited budgets.
• Complexity: 3D facial recognition systems can be more complex to develop and maintain than 2D systems, which can make them more difficult to implement and use.
• Data privacy concerns: Like other facial recognition technologies, 3D facial recognition systems raise concerns about data privacy and the potential for misuse of personal information. For example, there are concerns about the potential for 3D facial recognition to be used for surveillance or to track individuals without their consent.
• Bias and accuracy issues: There is a risk that 3D facial recognition systems, like other facial recognition technologies, may exhibit bias against certain groups of people, such as people of color or those with certain facial features. There may also be accuracy issues, particularly if the system is not trained on a diverse set of faces.
• Acceptability: Some people may find 3D facial recognition systems to be intrusive or uncomfortable to use, particularly if the systems require close proximity or contact with the face. This could limit the adoption of 3D facial recognition in certain setting.

D. Paper 4 Attendance Marking Using RFID

1. Identify the requirements and goals of the attendance system. This might include determining the size and location of the facility, the number of employees or students who will be using the system, and the level of security and accuracy required.
2. Choose and purchase the necessary hardware and software. This might include RFID readers, RFID tags or cards, and a computer or server to store and process the attendance data.
3. Install the RFID readers and configure the software. This might involve setting up the RFID readers at key locations around the facility, such as entrances and exits, and configuring the software to record and process attendance data.
4. Enroll employees or students in the system. This might involve creating records for each person in the database and assigning them a unique RFID tag or card.
5. Test the system to ensure it is functioning correctly. This might involve having employees or students use the system to clock in and out, and verifying that the attendance data is being accurately recorded and processed.
6. Train employees or students on how to use the system. This might involve providing instructions on how to use the RFID tags or cards to clock in and out, and addressing any questions or concerns they may have.
7. Roll out the system for use. This might involve activating the system for all employees or students, and setting up any necessary reporting or alert systems to monitor attendance data.
8. Monitor and maintain the system. This might involve regularly checking the hardware and software to ensure they are functioning correctly, and making any necessary updates or repairs.

a. Advantages

• Accuracy: RFID-based attendance systems can be more accurate than manual methods, because they use automated data collection and processing. This can reduce the risk of errors or omissions in the attendance data.
• Speed: RFID-based attendance systems can be faster to use than manual methods, because employees or students can simply swipe their RFID tags or cards to clock in and out. This can save time and improve efficiency.
• Convenience: RFID-based attendance systems can be convenient for both employees and managers, because they eliminate the need for manual data entry or paper sign-in sheets. This can reduce the workload for managers and make it easier for employees to track their attendance.
• Security: RFID-based attendance systems can offer improved security, because they use unique identification tags or cards to track attendance. This can make it more difficult for unauthorized individuals to access the system or for employees to clock in for each other.
• Reporting and analysis: RFID-based attendance systems can generate detailed reports and analytics about attendance data, which can be useful for tracking employee or student attendance patterns and identifying trends or issues.
• Flexibility: RFID-based attendance systems can be flexible, because they can be easily customized to meet the specific needs of an organization. For example, the system can be configured to allow employees to clock in and out from multiple locations, or to track attendance for different departments or teams separately.

b. Disadvantages

• Cost: RFID-based attendance systems can be more expensive to implement and maintain than other attendance tracking methods, because they require specialized hardware and software. This can be a significant barrier for organizations with limited budgets.
• Complexity: RFID-based attendance systems can be complex to set up and maintain, particularly for organizations that are new to the technology. This can require a significant investment of time and resources.
• Data privacy concerns: RFID-based attendance systems can raise concerns about data privacy, because they involve the collection and storage of personal information, such as employee or student names and attendance data. This can be a concern for individuals who are concerned about the potential for their data to be misused or shared without their consent.
• Dependence on technology: RFID-based attendance systems rely on technology to function, which can be a disadvantage if the system experiences technical issues or failures. This can disrupt attendance tracking and create inconvenience for employees or students.
• Accuracy issues: There is a risk that RFID-based attendance systems may not accurately track attendance in all circumstances. For example, the system may not register an employee's attendance if their RFID tag or card is not properly scanned.
• Acceptability: Some people may object to the use of RFID-based attendance systems, either because they are uncomfortable with the technology or because they view it as an invasion of privacy. This could lead to resistance or backlash from employees or Students

III. PROPOSED METHODOLOGY

The system proposed in the basis of face recognition. When a student come across the camera module, then his/her image/photo will be captured and recognize with validation. When recognition and validation is succeeded, then his/her attendance will mark automatically. In this system, user gets a login interface to interact with the system.

, interface displays the home page of the proposed system. The proposed block diagram of the automatic attendance system is shown in the Fig

  IV. TECHNOLOGIES USED

1. LBPH (Local Binary Pattern Histogram) is a Face-Recognition algorithm it is used to recognize the face of a person. It is known for its performance and how it is able to recognize the face of a person from both front face and side face.
2. Before starting the intuition behind the LBPH algorithm, let’s first understand a little bit about the basics of Images and pixels in order to understand how images are represented before we start the content about Face-Recognition. So let”s get started understanding images and pixels.

A. Working

The Local Binary Pattern (LBP) algorithm is a method used for texture classification in computer vision. It works by comparing each pixel in an image to its neighbors, and creating a binary code based on whether each neighbor is greater than or less than the pixel value. The resulting codes for all pixels in an image are then used to create a histogram, which can be used to represent the texture of the image.

The LBP algorithm has a number of variations, such as the Multiresolution Gray-Scale and Rotation Invariant Texture Classification (MG-RI-LBP) and the Spatial Gray-Scale and Rotation Invariant Texture Classification (SG-RI-LBP). These variations aim to improve the robustness and accuracy of the LBP algorithm.

In practice, the LBP algorithm is often used in combination with other techniques, such as support vector machines (SVMs), to classify images based on their texture. It has been applied in a wide range of applications, including face recognition, fingerprint identification, and terrain classification.

B. Design

GUI will be Web application

1. Face Recognition

a. Face model Training

• Load face image dataset
• Detect face region from images using open cv
• Resize images
• Train Face recognition model on face images and labels
• Save trained model

b. Testing

• Load image to test
• Load saved model
• Detect face region from image using open cv
• Resize image
• Test image using trained model
• Show predicted result

2. Admin module

a. Login with username and password

b. Register Student with basic details (username is the name given at the time of face model training) and also adding parent details with email

c. View Student details

d. View attendance details

e. Add Subjects  and time for attendance

3. Student module

a. View Attendance

b. Mark Attendance

• Input class room image
• Detect faces in image using  open cv
• Crop detected faces
• Resize images
• Input images to trained model
• If predicted result equal to the Student username, update attendance in database
• Attendance will be marked on the basis of the start time and end time of a subject (a single class is considered for testing).

  This admin panel contains:

1. Login Bar: This is a login interface where administrators can enter their credentials to access the admin panel.
2. Add Student: Once logged in, the admin can add student information such as name, ID, and other relevant details to the system.
3. Student Management: This section of the admin panel likely provides functionality for managing student records, such as editing or deleting student information.
4. View Attendance: The admin can access this section to view the attendance records of students. It may display attendance data in a tabular or graphical format.
5. Mark Attendance: Once the face recognition process verifies the student's identity, the attendance will be marked
6. Database: The attendance information is stored in a database, likely linked to the student's profile, capturing details such as the date and time of attendance.
7. View Attendance Panel: This section of the student panel enables students and admins to view their own attendance records.

VII. FUTURE ENHANCEMENTS

Integration with Student Information Systems: Integrate the automated attendance marker system with existing student information systems or learning management systems, enabling seamless data synchronization and generating comprehensive attendance reports and analytics.

Privacy and Security: Implement robust privacy and security measures to protect the collected biometric data and ensure compliance with data protection regulations. This can involve encryption, secure storage, and access control mechanisms.

Real-World Testing and Validation: Conduct extensive real-world testing and validation of the system in different environments, including classrooms, lecture halls, and large-scale events, to assess its performance and robustness in various scenarios.

Continuous System Improvement: Continuously monitor and evaluate the system's performance, gather feedback from users, and implement iterative improvements to enhance accuracy, reliability, and usability based on practical experiences and evolving technology advancements.

VIII. ACKNOWLEDGEMENT

We are very thankful to the Department of Computer Science and Engineering of Adi Shankara Institute of Engineering and Technology for permitting us to work on the topic “Automated attendance marker using facial recognition”. We truly express our gratitude to Prof. ANILA S, Department of CSE, ASIET for giving constant support and guidance.

Conclusion

In conclusion, the automated attendance marker system that utilizes the LBPH (Local Binary Patterns Histograms) algorithm offers an efficient and reliable approach for managing attendance based on face recognition. By following a series of steps including data collection, preprocessing, feature extraction, training, recognition, and matching, the system can accurately identify individuals and mark their attendance automatically. The LBPH algorithm extracts local texture patterns from preprocessed facial images, converting them into binary codes that form feature vectors for each face. These feature vectors are used for training the algorithm, enabling it to learn and differentiate between different individuals. During recognition, the system captures face images, preprocesses them, and extracts feature vectors for comparison with stored face templates. By matching the feature vectors and applying a confidence threshold, the system can accurately identify enrolled students and mark their attendance accordingly. The utilization of the LBPH algorithm in the automated attendance marker system provides several advantages, including its robustness to variations in lighting conditions, facial expressions, and angles. Additionally, the algorithm\'s simplicity and efficiency make it suitable for real-time attendance tracking applications. Overall, the automated attendance marker system based on the LBPH algorithm enhances efficiency, accuracy, and convenience in managing attendance processes, minimizing manual effort, and ensuring reliable attendance records.

References

[1] \"A Face Recognition Approach for Automated Attendance Management System\" by M. S. Kharate and S. K. Kopparapu. (IEEE International Conference on Computer, Communication and Control, 2015) [2] \"Automated Attendance Management System using Face Recognition\" by R. R. Yadav and S. S. Chaudhari. (IEEE International Conference on Advanced Computing and Communication Systems, 2016) [3] \"Automated Attendance System using Face Recognition\" by H. M. Meena, A. K. Tiwari, and S. Mishra. (IEEE International Conference on Computing, Communication, and Automation, 2018) [4] \"Automated Attendance System using Face Recognition Technique\" by S. R. Raut and S. B. Mane. (IEEE International Conference on Advanced Computing Technologies and Applications, 2019) [5] \"Automatic Attendance Management System using Face Recognition\" by N. S. Mirza, M. Z. Chishti, and M. Rafiq. (IEEE International Conference on Computer, Communication and Electrical Technology, 2016) [6] \"Face Recognition based Attendance Management System using PCA and DWT\" by S. Choudhury and M. Dey. (IEEE International Conference on Electrical, Electronics, Signals, Communication and Optimization, 2017) [7] \"Automated Attendance System using Face Recognition with Multiple Classifiers\" by S. P. Mathew and B. Chittilappilly. (IEEE International Conference on Recent Advances in Information Technology, 2019) [8] \"Face Recognition based Automated Attendance System using Eigenface and SVM\" by A. Verma and P. Yadav. (IEEE International Conference on Computational Intelligence and Computing Research, 2016) [9] \"An Automated Attendance System using Face Recognition\" by A. K. Jena and B. K. Mishra. (IEEE International Conference on Electrical, Electronics, Signals, Communication and Optimization, 2017) [10] \"Automated Attendance System using Face Recognition with Cloud Computing\" by K. D. Garde and S. K. Dhavale. (IEEE International Conference on Computer, Communication and Computational Sciences, 2017)

Copyright

Copyright © 2023 TN Krishnan Embranthiri, Maxin Shajan , Joepaul Jose , Prof. Anila S. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.