Comprehensive approach of real time web-based face recognition system using Haar Cascade and LBPH algorithm
Ayush Kumar, Deepanshi Singh · 2023
The goal of our research is to propose a system that automatically marks the attendance among the students in class using a webcam with Face Recognition. Since traditional ways of taking attendance are not that useful as it is highly time-consuming to mark the attendance of every student by calling their names as well as the use of papers to mark the attendance also affects our environment, therefore our system helps to save time as well as its eco-friendly as no paper is needed, helps to solve problems of fake attendance and proxies. The Haar Cascade & LBPH algorithms are used in this work to develop a Student’s Attendance System with facial recognition technology. Face detection and recognition are two steps in realtime human face recognition. As of Haar Cascades algorithm’s excellent precision and high real-time permit rate, which is implemented in OpenCV via the Python language, we use it for face detection. The training phase and evaluation phase are two categories taken into consideration in face recognition. The algorithm is trained during the training phase using samples of the image to be learned and the test image is compared across all trained samples in the dataset during the estimate step. By using Local Binary Patterns Histogram(LBPH), the live stream’s faces are recognized & the face characteristics are taken out. Face recognition is completed using the Euclidean distance classifier using the LBPH. The control of the system in computer-based communication is significantly impacted by authentication. The system will automatically keep track of students' attendance in a classroom and will give faculty members the ability to conveniently access student data by keeping a log of their clock-in times.