Real Time Pattern Recognition with Support Vector Machines and Local Binary Patterns

This research introduces a facial recognition-based attendance system that leverages an open-source computer vision library and integrates with a real-time database system. The system comprises essential components, including a camera for image capture, a computer for image processing, and a database containing registered facial data. As individuals enter a room, the camera captures their images, which are then subjected to facial detection and recognition by the computer. The system conducts a comparison between the detected faces and the pre-registered facial data in the database, subsequently logging the attendance of individuals who are successfully recognized. Moreover, the system offers functionality for the inclusion of new facial data into the database, facilitating future attendance tracking. To gauge the system's performance, an assessment was carried out using a dataset of 100 images. The results indicate an accuracy rate of 95% in both facial detection and recognition tasks. This system holds promise for a wide range of practical applications in diverse environments, such as educational institutions, workplaces, and event management, by automating attendance management processes and alleviating the burdens associated with manual record-keeping.