Automated Smart Attendance System with Face Recognition

In this paper, the development of the smart attendance system with face recognition technology is presented. The proposed system is fully automated and does not require. The system has been implemented on a Raspberry Pi 4 Model B board using a USB webcam automatically detects and recognizes people in real-time and records their presence in a centralized database. The proposed system utilizes OpenCV and the concept of face recognition models like CNNs to guarantee proper recognition regardless of changes lighting conditions. When compared with the traditional method, this contactless technology has been found to result in a lower increase in errors and prohibits proxy attendance. Results for the experimental evaluation of the hardware prototype justify the following conclusions on the desirability and functionality an average face recognition accuracy of over 90% (95% in well-lit conditions, around 90% in low-lighting conditions) and average processing latency of approximately 1-2 seconds per recognition. The power consumption is restricted to a standard *5V/3A USB Power Supply, for continuous use*. The paper describes the hardware and software components "Software design, methods of software testing, including unit and black box tests, and performance measurement." Lastly, the benefits of the system are highlighted, and future enhancements are touched upon