An Ensemble for Attendance Management with Face Visualization and Recognition Using Local Binary Pattern Histogram

Abstract

Identifying a person and recognizing their individuality has been the most important necessity in any community or organization. So, attendance management is a necessary tool in any environment where maintaining attendance is required. Automation is always an easier way to upgrade the existing approach, in performance, efficiency, and robustness. This project is aimed at developing a less intrusive, cost-effective, and more efficient automated attendance management system. The system is based on face detection and recognition algorithms, which automatically detects a person through a walk-in or an image of an individual or a group. The system comports the given image with the pre-registered data and confronts the lateral procedure. The system uses Local Binary Pattern Histogram Algorithm with Haar cascade classifier to identify the image through its nodal preference to downscale, detect and differentiate the image. The produced data of the attendance is managed and automatically updated in the database and is stored in an excel sheet. Different real-time scenarios were taken into account to evaluate the performance of the system. The project also has measures implemented to handle the threats like spoofing.