Abstract: Deep learning has been revolutionizing the field of artificial intelligence in recent years, with significant advancements in computer vision, natural language processing, speech recognition, and many other areas. Some of the most commonly used deep learning techniques and architectures include convolutional neural networks (CNNs). Till now there are Several issues that can arise when using deep learning algorithms for face recognition for attendance, the accuracy of face recognition systems depends heavily on the quality of the input data. low lighting conditions, occlusions, and changes in facial hair or makeup. These limitations can impact the accuracy of the system and result in incorrect attendance records. The existing system captures images of students and compares them to a database of registered images to mark attendance. However, the system suffers from several drawbacks, such as low accuracy in identifying students with different facial expressions, lighting conditions, or angles. To address these issues. This project proposes a face recognition system for attendance in educational institutions that utilizes deep learning algorithms and computer vision techniques. The system works by capturing the image of the student's face using a camera. The captured image is then processed by the gray scale and CNN algorithm to detect and recognize the student's face. If the system recognizes the student's face, their attendance is marked as present. The proposed system has several advantages over traditional attendance management systems. It eliminates the need for manual attendance taking, which saves time and reduces the errors. It also provides real-time attendance information, which can be useful for monitoring attendance patterns and identifying students. The system has been implemented and tested on a dataset of student image, achieving an accuracy rate of 95%. The system is scalable and can be easily integrated into existing educational institution infrastructure.
Keywords: Face recognition system for educational institutions that uses the Convolutional Neural Network (CNN)algorithm, deep learning, student dataset, Gray scale, Haar cascade, data registration.
| DOI: 10.17148/IJARCCE.2023.124176