Attendance Marking System using Facial Recognition

Abstract

Marking attendance is of significant importance in many places such as classrooms, conferences, or even prisons. It is possible to have a system (application) which marks attendance of people with least possible human intervention.This paper discusses on the face recognition methodologies using the concepts of image processing, computer vision and machine learning. The system is implemented using two different feature extraction techniques namely Histogram of Oriented Gradients (HOG) and Principle Component Analysis (PCA). The HOG features are classified using Support Vector Machine (SVM) classifier. Euclidean distance is used for classification in the case of PCA. The process of face detection is performed using Viola-Jones algorithm. Real-time or live facial recognition is also possible in the said system using HOG feature extraction and SVM classifier. The face detection and tracking in real-time face recognition is performed using Viola-Jones and Kanade-Lucas-Tomasi (KLT) algorithm respectively.