THE 3 rd INTERNATIONAL SCIENTIFIC CONFERENCES OF STUDENTS AND YOUNG RESEARCHERS dedicated to the 99
th
anniversary of the National Leader of Azerbaijan Heydar Aliyev
166
FACE RECOGNITION AND ATTENDANCE CHECKING Ilgar Mammadov Baku Higher Oil School Baku, Azerbaijan ilgar.mammadov.std@bhos.edu.az Supervisor: PhD, Associate Professor Leyla Muradkhanli Keywords: face recognition, attendance checking system, deep convolutional neural
network, support vector machine
Abstract This thesis investigates an attendance checking system that is based on
face recognition algorithm. In this paper the theory behind the face recognition
is explained, its usage in attendance checking process is analyzed, and the
required software is explained.
Introduction In the modern world, face recognition system is used extensively and
solves many time-requiring problems rapidly. This system can be applied to
attendance checking process as well.
Even though some modernized methods are used for that purpose, they
also have some disadvantages. For example, in many places, attendance is
recognized by fingerprint sensors. However, the main inefficiency is the time
required for the attendees to wait in the queue. Additionally, one global
problem today is the pandemic illnesses and pressing finger to the same sensor
one-by-one increases the risk.
Instead, broad usage of face-recognition systems for taking the attendance
can lead to the more efficient time-control and reduced risk of infectious
diseases.
Face recognition algorithm The steps that are performed by the algorithm are as follows:
1. Capture an image with a camera
2. Convert the image to grayscale
Compare the intensities of pixels and generate arrows (gradients) to
differentiate darker pixels. The result will be
histogram of oriented gradients image (HOG) (Figure 1.) :
Figure 1.
Histogram of oriented gradients image (HOG)