Smart Attendance Using Face Recognition

Abstract

Face recognition is one of the greatest image-handling techniques, which is critical in the specialized field. The identification of the human face is a practical problem for validation purposes, particularly when understudies are involved. A face biostatistics-based participation framework is a strategy for identifying pupils based on high-quality observation and other PC developments. This framework's development intends to digitalize the old way of gauging participation by calling names and keeping pen-and-paper records. In this digital era, every organization is concerned about recording the presence of someone's attendance is important. Generally, attendance is often done by using manually signed within the register, calling one by one, RFID, and a Biometric system. To enhance time efficiency and authentication this paper proposes a sensible attendance system using Machine learning algorithms to record the attendance from everyone present in an organization. The proposed system is meant for automating the attendance of the various organizations and reduces the issues of the prevailing manual system. Once faces are detected and recognized with the prevailing database, the system calculates attendance for the recognized persons with the respective ID in real-time, and an excel sheet generated and saved by the system automatically. To implement the smart attendance system (model) for a specific face and distinguish it from an outsized number of stored faces with some real-time variations also. It uses OpenCV and Python to get the dataset, record the attendance and store the end in the excel sheet. The proposed system is meant for automating the attendance of the various organizations and reduces the issues of existing manual system. Once faces are detected and recognized with the prevailing database, the system calculates attendance for the recognized students with the respective subject in real time. And an excel sheet generated and saved by the system automatically.