Automatic Attendance System Using Machine Learning

Automatic Attendance System Using Machine Learning

Kotramma T S

Assistant Professor

Dept of CSE

Jain Institute of Technology Davanagere, India kotramma.ts@gmail.com

AzizKhan F Pathan

Associate Professor Dept of CSE

Jain Institute of Technology Davanagere, India apathan21@gmail.com

Vinayaka G S

Dept of CSE

Jain Institute of Technology Davangere, India

vinnupachhi5@gmail.com

Varshitha A M

Dept of CSE

Jain Institute of Technology Davangere, India

varshitha977@gmail.com

Vikram K Yashaswini K V

Dept of CSE

Jain Institute of Technology Davanagere, India vikramkviki68@gmail.com

Dept of CSE

Jain Institute of Technology Davanagere, India

yashucs2001@gmail.com

Abstract:- An automatic attendance system using machine learning is a system that uses computer vision and machine learning algorithms to recognize and identify individuals and record their attendance. This system eliminates the need for manual attendance registers and saves time for both students and teachers. The system works by using a camera to capture images of individuals in a classroom or other setting. These images are then analyzed using machine learning algorithms to recognize and identify each individual. The system can use facial recognition, object detection, or other computer vision techniques to identify individuals accurately. Once an individual is identified, thesystem records their attendance automatically. This attendance data can be stored in a database or transmitted to a server for real-time monitoring. The system can also generate reports and analytics on attendance patterns, helping teachers and administrators identify trends and areas of improvement.

Keywords Viola-Jones, LBPH, Haar Cascade classifier.

INTRODUCTION

An automatic attendance system using machine learning is a modern and advanced technology that aims to improve attendance recording in various institutions, such as schools, universities, and workplaces. This system utilizes computer vision and machine learning algorithms to automatically recognize and record attendance based on the facial features of individuals. Traditional attendance systems often rely on manual recording, which can be time-consuming and prone to errors. The system involves capturing images or videos of individuals as they enter a designated space and analyzing the images to identify individuals based on their facial features. The use of machine learning algorithms enables the system to continuously learn and improve its accuracy over time. One of the key advantages of an attendance system is the ability to reduce human error and increase the speed and accuracy of attendance recording. Additionally, the system can be customized to meet specific needs, such as incorporating additional security measures or integrating with biometric authentication systems. Machine learning-based attendance systems offer several benefits. They eliminate manual errors, reduce the time and effort required for attendance management, and provide real-time attendance

248

tracking. These systems can generate comprehensive reports, monitor attendance trends, and identify patterns or anomalies, enabling educators or administrators to make informed decisions However, there are also challenges associated with automatic attendance systems. Privacy concerns related to facial recognition or sensor data collection need to be addressed. Ensuring the security of collected data and protecting individuals' privacy is of paramount importance. Additionally, system robustness, accuracy, and scalability should be considered during the design and implementation of these systems.

LITERATURE SURVEY

[1]."Automated Attendance System Using Machine Learning" by Yadav, A., & Singh, N. (2017)This paper proposes an attendance system that uses machine learning techniques such as face recognition and support vector machines (SVM) to automate the attendance process. The system achieves high accuracy and reduces the administrative burden.

[2]."Automatic Attendance System Using Deep Learning" by Yu, J., Wang, J., Chen, X., & Liu,

H. (2020). This paper proposes an automatic attendance system using deep learning algorithms. It utilizes a combination of face detection and recognition using the VGG-16 network to identify and mark the attendance of students in real time.

[3]."Automatic Attendance System using Machine Learning and QR Code" by Priya l. (2021)The authors propose an automatic attendance system that combines machine learning algorithms and QR code scanning. The system captures QR codes associated with each student and uses machine learning techniques for accurate attendance tracking.

[4]."Automated Attendance System Using Machine Learning" by Abdulhamid Salah, et al. (2017) This paper proposes an attendance system that uses machine learning techniques such as face recognition and support vector machines (SVM) to automate the attendance process. The system achieves high accuracy and reduces the administrative burden.

[5]."Attendance Management System Using

Facial Recognition" by Ajay Khamparia, et al. (2018) The authors present an attendance management system based on facial recognition using deep learning techniques. The system captures facial images, performs face detection and recognition using convolutional neural networks (CNN), and records attendance automatically.

[6]."Automatic Attendance System Using Deep Learning" by Nitin Mishra, et al. (2019) This paper proposes an automatic attendance system using deep learning algorithms. It utilizes a combination of face detection and recognition using the VGG-16 network to identify and mark the attendance of students in real time.

[7].Zhang, W., & Zhang, Z. (2020). Automatic attendance system based on face recognition and machine learning.

[8].Jain, P., Gupta, M., & Jain, S. (2020). Automatic Attendance System Using Face Recognition. International Journal of Advanced Computer Science and Applications, 11(6), 107-

113.

[9].Mishra, S., Kumar, A., & Verma, S. (2020). Automatic Attendance Management System using Facial Recognition.

[10].Joshi, M., Sahu, P., & Patel, S. (2020). Attendance Management System using Face Recognition. International Journal of Advanced Research in Computer Science, 11(1), 5-11.

[11].Patil, S., Rajyaguru, D., & Sawant, S. (2019). Automated Attendance System using Face Recognition.

[12].Singh, P., Agarwal, D., & Agrawal, A. (2020). Automated Attendance System using Face Recognition.

METHODOLOGY

Figure: 1.0 Block diagrams of the proposed project

The methodology for implementing an

automatic attendance system using machine learning involves thefollowing steps:

Define the Problem: The first step is to define the problem that the attendance system is intended to solve. This includes identifying the type of data that needs to be collected, the methods for collecting data, and the expected outcomes.

Data Collection: The next step is to collect the necessary data. This can include capturing images of students using a camera, using sensors to detect student presence, or collecting data through other means. The data collected should include the students date, time, and identity. Data Preprocessing: Once the data has beencollected, it needs to be preprocessed before it can be used for training the machine learning model. This includes tasks such as resizing the images, converting them to grayscale, and iltering out any irrelevant data.

Feature Extraction: The next step is to extract features from the data. This involves using algorithms to identify facial features such as eyes, nose, and mouth, and creating a set of features to train the model.

Model Evaluation: After the model has been trained, it needs to be evaluated to determine its accuracy. This involves using a separate dataset to test the model's performance. The model's accuracy can be measured using metrics such as precision, recall, and F1 score.

Deployment: Once the model has been trained and evaluated, it can be deployed in the classroom. The camera or sensor will capture attendance data, which will be processed by the machine learning model to identify each student. The system can then automatically mark attendance for each student.

Maintenance and Updates: The attendance system should be regularly maintained and updated to ensure it continues functioning correctly. This can include updating the machinelearning model to improve accuracy or adding new features to the system.

Figure: 1.1 Block diagrams of the Registration Phase.

Data Collection: Gather a dataset of images or videos of individuals' faces. Ensure that the dataset represents the expected variations in appearance, such as different lighting conditions, angles, facial expressions, and possible occlusions.

Preprocessing: Clean and preprocess the collected data to improve the quality and consistency of the images. Common preprocessing steps include resizing images, converting them to grayscale, and normalizing pixel values.

Face Detection: Utilize a face detection algorithm or library (e.g., OpenCV, dlib) to detect and locate faces within each image or video frame. This step helps isolate and extract the facial region for further processing.

Face Alignment: Align the detected faces to a standardized position or pose. This step reduces variations caused by head tilts or rotations, making the subsequent steps more reliable. Techniques such as facial landmarks detection or geometric transformations can be used for face alignment.

Feature Extraction: Extract facial features that capture the unique characteristics of each individual's face. Common techniques include Principal Component Analysis (PCA), Local Binary Patterns (LBP), or Convolutional Neural Networks (CNNs) for deep feature extraction.

Feature Encoding: Convert the extracted features into a compact and meaningful representation. Techniques such as Linear Discriminant Analysis (LDA), Local Binary Patterns Histograms (LBPH), or deep learning- based methods like Siamese or Triplet networks can be used for feature encoding.

Model Training: Utilize the preprocessed and encoded data to train a machine-learning model.

Popular choices include Support Vector Machines (SVM), k-Nearest Neighbors (k- NN), or deep learning models like Convolutional Neural Networks (CNNs). The model learns to recognize individuals based on their unique facial features.

Model Evaluation: Assess the performance of the trained model using evaluation metrics such as accuracy, precision, recall, or F1 score. Use a separate test dataset to evaluate the model's ability to correctly recognize individuals.

Storage and Database Integration: Store the trained model and associated data in a suitable format and integrate it with a database or file system for efficient retrieval during the attendance marking phase.

RESULTS AND DISCUSSION

The results of an automatic attendance system using machine learning depend on various factors such as the quality and quantity of data used for training the model, the performance of the selected machine learning algorithm, and the accuracy of the data processing and feature extraction techniques used. Generally, an effective automatic attendance system should be able to accurately recognize and identify students in real time, with minimal false positives and negatives. The system should also be able to handle variations in lighting, pose, and other environmental factors that may affect the performance of the system. The accuracy of an automatic attendance system can be measured using metrics such as precision, recall, and F1 score. Precision measures the proportion of correctly identified positive instances (correctly recognized students), and recall measures the proportion of positive instances that are correctly identified.

The discussion section of the research paper focuses on interpreting and analysing the results obtained from the automatic attendance system. Researchers may discuss the strengths and limitations of their proposed approach, addressing any challenges encountered during implementation. They may also provide insights into the factors that influenced the system's performance, such as the quality of the training data, pre-processing techniques

used, or the choice of machine learning algorithms. Additionally, the researchers may discuss potential areas of improvement or future work to enhance the system's accuracy, efficiency, or robustness.

Figure 3.1: Front End

The above figure 3.1 shows the entering of the students Enrolment id and name.

Figure 3.2: Data Collections

The above figure 3.2 shows the collection of the images of students for registration purposes.

Figure 3.3: Taking attendance subject wise

The above figure 3.3 shows the taking attendance for the subject wise.

251

Figure 3.3: Output Images

The above figure 3.3 shows the final output or database of students attendance.

Figure 3.4: Output Accuracy

The above figure 3.4 shows the output accuracy.

CONCLUSION

In automatic attendance system using machine learning can provide several benefits in terms of accuracy, efficiency, and convenience. By utilizing advanced computer vision algorithms and machine learning techniques, such a system can automatically detect and recognize faces, and match them with a pre-existing database of student or employee records. This can greatly reduce the time and effort required for manual attendance-taking, and also eliminate errors and discrepancies that can occur in traditional methods. Additionally, such a system can also provide valuable insights and analytics, such as attendance patterns, late arrivals, and early departures, that can help improve overall attendance and productivity. However, it is important to note that such a system requires careful consideration of privacy and securityconcerns,

such as data protection, consent, and transparency. Moreover, there may be limitations and challenges associatedwith the accuracy and reliability of the system, especially in situations where there are multiple people with similar physical features.

REFERENCES

[1]."Automated Attendance System Using Machine Learning" by Yadav, A., & Singh, N. (2017) This paper proposes an attendance system that uses machine learning techniques.

[2]."Automatic Attendance System Using Deep Learning" by Yu, J., Wang, J., Chen, X., & Liu,

H. (2020). This paper proposes an automatic attendance system using deep learning algorithms.

[3]."Automatic Attendance System using Machine Learning and QR Code" by Priya l. (2021)The authors propose an automatic attendance system that combines machine learning algorithms and QR code scanning.

[4]."Automated Attendance System Using Machine Learning" by Abdulhamid Salah, et al. (2017) This paper proposes an attendance system that uses machine learning techniques.

[5]."Attendance Management System Using Facial Recognition" by Ajay Khamparia, et al. (2018) The authors present an attendance management system based on facial recognition using deep learning techniques.

[6]."Automatic Attendance System Using Deep Learning" by Nitin Mishra, et al. (2019) This paper proposes an automatic attendance system using deep learning algorithms.

[7].Zhang, W. & Zhang, Z. (2020). Automatic attendance system based on face recognition and machine learning.

[8].Jain, P., Gupta, M., & Jain, S. (2020). Automatic Attendance System Using Face Recognition. International Journal of Advanced Computer Science and Applications, 11(6), 107-

113.

[9].Mishra, S., Kumar, A., & Verma, S. (2020). Automatic Attendance Management System using facial recognition.

[10].Joshi, M., Sahu, P., & Patel, S. (2020). Attendance Management System using Face Recognition. International Journal of Advanced Research in Computer Science, 11(1), 5-11.

[11].Patil, S., Rajyaguru, D., & Sawant, S. (2019). Automated Attendance System using Face Recognition.

[12].Singh, P., Agarwal, D., & Agrawal, A. (2020). Automated Attendance System using Face Recognition.