You are on page 1of 3

Face Recognition based Attendance System

Abstract

The attendance system is used to keep track of whether or not a


student shows up for class. Biometric-based, Radiofrequency card-based,
face recognition-based, and classic paper-based attendance systems are
all examples of distinct types of attendance systems. Face recognition-
based attendance systems are the most secure and time-saving of them all.
There are various study studies that focus solely on student recognition
rates. This study focuses on a face recognition-based attendance system
that has a lower false-positive rate by employing a confidence threshold,
such as the euclidean distance value, to discover unknown people and
save their photographs. The Local Binary Pattern Histogram (LBPH)
approach outperforms previous euclidean distance-based algorithms like
Eigenfaces and Fisherfaces. Because of their robustness, Haar cascade
was utilised for face detection, while the LBPH algorithm was used for
face recognition. It withstands monotonic grayscale modifications with
ease. This system is evaluated utilising scenarios such as facial
recognition rate, false-positive rate for that, and false-positive rate with
and without a threshold in detecting unfamiliar persons. Students are
recognised by this technique even if they are wearing glasses or have
grown a beard.
Introduction

Automatic face identification is an essential theme of computer


vision and pattern recognition analysis. Control of machines and their
process with various technologies based on computer software is called
automation.Traditional attendance systems like paper-based method of
marking attendance is time-consuming and its complexity increases with
the increase of overall strength. This case is nullified in this automated
version has it saves time and an additional bonus comes with security as
it also helps to prevent proxy of attendance. Here used Haar cascade for
face detection because of their robustness and LBPH algorithm for face
recognition. This system even detects and saves the images of any
unknown person in the class whose information is not there in the
database.

Project Flow
 Data collection
 Pre-processing
 Build face recognizer
 Training
 Save trained model
 Face recognition

Methodology

The first step is to collect data. Different facial photos will be used to collect
data. Using the LBPH face recognizer, the collected data will be pre-
processed and trained. This is how a model is made. This model will be able
to recognise the face that has been taught to it. If we use the same face as
input, the model will be able to identify the person.

Software Requirements
1. Anaconda
2. Spyder - IDE
3. Python - programing language

You might also like