2022 12th Iranian/Second International Conference on Machine Vision and Image Processing (MVIP)
Department of Electrical and Computer Engineering, Shahid Chamran University of Ahvaz, Iran, 23 & 24 February 2022
978-1-6654-1216-2/22/$31.00 ©2022 IEEE
FFDR: Design and implementation framework for
face detection based on raspberry pi
Dhafer Alhajim
Department of Computer Technical
Engineering
Technical Engineering college,
The Islamic University, 54001
Najaf, Iraq
idcte202006@iunajaf.edu.iq
Gholamreza Akbarizadeh
Department Of Electrical Engineering,
Faculty Of Engineering,
Shahid Chamran University Of Ahvaz,
Ahvaz, Iran
g.akbari@scu.ac.ir
Karim Ansari-Asl
Department Of Electrical
Engineering,
Faculty Of Engineering,
Shahid Chamran University Of Ahvaz,
Ahvaz, Iran.
karim.ansari@scu.ac.ir
Abstract— In today's world, we are surrounded by data of
many types, but the abundance of image and video data
available offers the data set needed for face recognition
technology to function. Face recognition is a critical component
of security and surveillance systems that analyze visual data and
millions of pictures. In this article, we investigated the possibility
of combining standard face detection and identification
techniques such as machine learning and deep learning with
Raspberry Pi face detection since the Raspberry Pi makes the
system cost-effective, easy to use, and improves performance.
Furthermore, some images of a selected individual were shot
with a camera and a python program in order to do face
recognition. This paper proposes a facial recognition system that
can detect faces from direct and indirect images. We call this
system FFDR, which is characterized by high speed and
accuracy in the diagnosis of faces because it uses the Raspberry
Pi 4 and the latest libraries and advanced environments in the
Python language.
Keywords— facial recognition, machine learning, deep
learning
I. INTRODUCTION
The Raspberry Pi is a miniaturized, card-sized computer
consisting of a conventional keyboard and mouse that
connects to a computer display or TV. The most recent version
of this little board includes certain new capabilities that make
it capable of replacing desktop PCs [2]. The previous versions'
one-size-fits-all approach is gone with the Raspberry Pi 4. It
comes with one, two, or four gigabytes of RAM. (This is the
first time a Pi with more than 1 GB of RAM has been
available.) The additional RAM expands the Pi's capabilities,
including the ability to execute software, but the Raspberry Pi
4 remains the same excellent small DIY gadget. The
Raspberry Pi started as a hacker's dream: a low-cost, low-
power, highly extensible, hackable PC that came in a bare
circuit board form factor [4]. It has been used to power
everything from scaled-down Mars rovers to millions of
science and hacking day projects in schools throughout the
world [5]. It was designed as a one-part instructional gadget
and a one-part tinkering tool. Face detection, often known as
facial recognition, is a computer technology that uses artificial
intelligence (AI) to discover and recognize human faces in
digital pictures. Face recognition technology may be used to
offer real-time monitoring and tracking of individuals in a
variety of sectors, such as security [6] and personal safety [7].
It has developed from basic computer vision approaches to
breakthroughs in machine learning (ML) and associated
technologies, with the consequence being continual
performance gains. It currently serves as a crucial first step in
a variety of vital applications, including face tracking, face
analysis, and facial identification. Face recognition has a
major effect on the application's ability to conduct consecutive
tasks. Face detection aids in determining which sections of an
image or video should be focused on in order to evaluate age,
gender, and emotions by utilizing facial expressions in face
analysis [8]. Face recognition software employs algorithms
and machine learning to locate human faces within bigger
pictures that frequently include non-face items, including
landscapes, buildings, and other human body parts such as feet
and hands [9]. Human eyes are one of the simplest features to
identify, so face detection algorithms usually start there. After
that, the algorithm could try to recognize the brows, mouth,
nose, nostrils, and iris. Once the system has determined that a
facial region has been discovered, it performs additional tests
to ensure that it has recognized a face [3]. The algorithms must
be trained on colossal datasets, including too many positive
and negative images, to achieve precision. The algorithms'
ability to detect whether or not there are faces in a picture and
where they are increases as a result of the training [10].
Because of the variety of elements such as orientation,
expression, posture, position, pixel values, and skin color, the
existence of spectacles or facial hair, and changes in camera
gain, lighting conditions, and picture resolution, detecting
faces in figures can be difficult. Face identification using deep
learning has advanced in recent years, with the benefit of
considerably outperforming standard computer vision
approaches [11]. Face recognition, in other words, goes
beyond detecting the existence of a human face to discovering
whose face it is. The process employs a computer application
that takes a digital image of a person's face, often obtained
from a video frame, and compares it to photos stored in a
database. In this paper, we will learn how to conduct face
recognition using OpenCV. To create our face recognition
system, we'll first conduct face detection, then use machine
learning to extract face embedding from each face, train a face
recognition system on the embedding, and lastly use OpenCV
to recognize faces in both images and video streams.