Face recognition based attendance system

Abstract

Face recognition systems have gained a considerable amount of attention as they can be used in many different applications. Applications where such systems are used include those for identification purposes in airports, in surveillance cameras to detect potential criminals, in computers and mobile phones as a form of biometric authentication, and in other forms of technology such as robotics. Face recognition systems work by comparing selected facial features from an image with faces in a training set.

The objective of this project is to develop a face recognition based attendance system for students to take their attendance. Students will take their attendance by looking into the camera for their attendance to be taken. This will help to save the teachers time and effort as teachers will not need to spend time to take the attendance of the students and the attendance of up to five students can be taken at a time.

Other sub-tasks in this project include creating the back-end portion of the system where administrators can also test the face recognition accuracy and face recognition speed of the system by adjusting the various settings. The various techniques used in face detection and face recognition such as the Viola Jones detection framework, Principal Component Analysis and Eigenfaces were also being explored. The factors which affect the face recognition speed and face recognition accuracy were also determined. Other tasks include finding out how super-resolution techniques were able to improve the resolution of the students’ faces to ensure that their faces can be recognized correctly. Experiments were also conducted to determine how the various settings affect the face recognition accuracy and face recognition speed. The face recognition application was developed using the C# programming language.