Face Detection Using Artificial Intelligence and Machine Learning with Python

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 10 Issue: 09 | Sep 2023 www.irjet.net p-ISSN: 2395-0072
© 2023, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 57
Face Detection Using Artificial Intelligence and Machine Learning with
Python
P.V.K.S Varun1, Syed.Tasleem2, R.Sivateja3 T.Dheeraj Reddy4
1Department of Electronics and Computer Science Engineering, Mahindra University, Hyderabad
2,3Department of Computer Science and Engineering, Narayana Engineering College, Gudur
4Department of Computer Science and Engineering, Mahindra University, Hyderabad
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The primary objective of this research paper
was to develop a Python-based solution, in conjunction with
the OpenCV computer vision library, for the detection and
recognition of faces in images. The major focus of this study
revolved around face detection and recognition. To achieve
face detection, the algorithm used here is Haar Cascade
algorithm. The advancements in artificial intelligence and
machine learning technologies in the modern era have
significantly elevated the world to new levels. Moreover,
many impossible circumstances which might be challenged
by human beings can be solved with the useful resource of
the brand new technologies along with artificial intelligence
and gadget gaining knowledge of artificial intelligence and
system gaining knowledge of have wide applications in
distinct fields as an example, computer vision and robotics.
Data is essential for machine learning and artificial
intelligence as well as in lots of initiatives. To understand
artificial intelligence actually, by enabling the unlocking of
devices like smartphones through facial recognition, this
paper showcases the practical application of artificial
intelligence. As a result, this paper offers a comprehensive
package that combines theoretical knowledge with real-
world implementations of artificial intelligence and
machine learning technologies
Key Words: Artificial intelligence, Data, Machine learning,
OpenCV, Python, Algorithm
1. INTRODUCTION
In this modern era advanced technologies have become
ubiquitous, encompassing our lives. With compact devices
no larger than the palm of a hand, AI applications grant us
access to vast amounts of information from all corners of
the globe. Through artificial intelligent software, human
life has been streamlined and simplified in numerous
ways. Moreover, the presence of self-learning algorithms
and the abundance of online data, coupled with affordable
computation, have propelled machine learning to
unprecedented heights. Artificial intelligence's popularity
has rapidly grown, becoming necessary in everydays life
and fostering hope for humanity's future. Advancements
in computing power and data handling have demonstrated
AI's success. Machine intelligence, surpassing linear
programming, allows computers to learn from inputs. AI,
an interdisciplinary science, imitates human behaviors
and simplifies life. Despite various definitions, AI
significantly benefits society on a broad scale. This project
focuses on face detection and recognition, a well-known
technology in computer vision and image processing. It
aims to detect human faces in digital images and videos,
whether it's a single face or multiple faces. The main
objective is to use artificial intelligence and machine
learning, along with Python programming and the OpenCV
library, to detect and recognize faces in images. The
project begins with detecting a single face and progresses
to recognition using the same facial image dataset,
identifying the person in the image.
1.1 About Artificial Intelligence
Artificial intelligence (AI) is the ability of computers or
machines to understand and learn tasks, resembling
human intelligence. It encompasses a wide range of
human-like behaviors, such as memory, emotion,
reasoning, and communication, which can be simulated by
computer systems[1]. AI allows machines to learn from
experience and perform tasks like problem-solving,
language comprehension, and medical diagnosis. AI
involves the development of computer systems capable of
performing tasks that typically require human intelligence,
such as visual perception, speech recognition, decision-
making, and language translation.
AI-powered machines possess the capability to analyze
vast amounts of data rapidly and with minimal errors
compared to humans[2]. By utilizing real-world data and
recognizing patterns within it, AI machines can
continuously learn and adapt to new situations. AI has
found applications in various fields, including computer
vision, machine learning, natural language processing,
game theory, cognition, reasoning, and robotics.
Advancements in hardware and software technologies
have propelled the AI field to new heights since the digital
revolution, the advent of personal computers, and the rise
of the internet.

1.2 History of AI
Artificial Intelligence (AI) has been a captivating subject in
computer science, studied for decades by scientists in
Engineering, Mathematics, and Computer Science,
exploring the concept of artificial brains. The term
"artificial intelligence" was coined in 1956 by John
McCarthy during the first conference on the subject. In
1942, Isaac Asimov introduced the three laws of robotics,
inspiring future AI and robotics researchers. The official
naming of AI occurred during the Dartmouth Conference
in 1956, where the first AI system, the logical theory
program, was introduced.
From the 1970s onward, AI expanded into machine
learning, expert systems, pattern recognition, and
robotics. AI's growth since the 21st century has been
remarkable, transforming human society and evolving
across various disciplines. Advancements in machine
learning, big data, cloud computing, and AI algorithms
have accelerated its development[3]. AI is now extensively
used in search engines, voice recognition, autonomous
vehicles, home appliances, and more. The future promises
even greater AI advancements, benefiting businesses,
healthcare, public sectors, and social interactions with the
increasing use of machine learning for predictive analytics.
2. ARTIFICIAL INTELLIGENCE AND MACHINE
LEARNING
Over the past few years, the terms "artificial intelligence"
(AI) and "machine learning" (ML) have become popular
term. While many people use these terms interchangeably,
it's important to understand that machine learning is a
subset of artificial intelligence. AI encompasses the
development of intelligent computer systems capable of
performing tasks traditionally requiring human
intelligence. It's a vast field, with machine learning being
one application that trains machines to learn from data and
experiences. In machine learning, computer systems use
various algorithms and statistical modules to perform
specific tasks without explicit programming. It plays a
crucial role in achieving artificial intelligence. The
application of AI and ML has grown significantly in various
sectors, with machine learning being utilized to solve real-
world problems by learning from data and making
decisions with minimal human intervention.
Data is a crucial element for both AI and machine learning.
Building systems that understand human conversations,
recognize images, or learn from surroundings requires
ample data. Access to large quantities of data has become a
defining factor for the success of AI-driven companies.
Deep learning is a subset of machine learning, specifically
referring to the use of deep artificial neural networks. It
has gained popularity due to its power and effectiveness in
various applications, such as sound recognition, image
recognition, natural language processing, and
recommender systems.AI is no longer confined to big
companies; it is prevalent in our daily lives. Features like
Face ID on smart phones and automatic voice assistants
demonstrate how AI is integrated into everyday
technology. AI has found applications in various industries,
including finance, travel, healthcare, retail, transportation,
journalism, education, agriculture, entertainment, and
government. For instance, AI improves customer service,
security, and advertising in the finance sector, simplifies
hotel bookings and airport check-ins in the travel industry,
and aids in medical diagnoses and patient care in
healthcare. AI's impact is also evident in the retail industry
through automatic payment machines, inventory
management, and cashier-less shopping experiences. In the
gaming industry, AI is used to create intelligent agents
capable of challenging human players and continually
improving their strategies.
Overall, AI is rapidly changing our society and various
industries, leading to advancements in the smart education
system, agriculture, journalism, entertainment, and more.
Its widespread adoption is transforming everyday life, the
economy, and society as a whole.
Machine learning is an AI subfield that designs algorithms
for computers to learn from data and experiences without
explicit programming[4][5]. It enables AI to make
predictions, identify trends, and understand changes in
fields like finance and medicine. Reinforcement learning
involves trial and error to create consistent policies, as
seen in AI game strategies[6][7]. Machine learning finds
diverse applications in healthcare, automotive, cyber
security, and more. Real-life examples include virtual
assistants, dynamic pricing, email filtering, and fraud
detection.
3. DETECTION AND RECOGNITION OF FACE
USING PYTHON
Face detection and recognition stand out as prominent
computer vision technologies in the realm of artificial
intelligence due to their wide-ranging utility. Face
recognition involves identifying individuals by analyzing
facial attributes through various methodologies[8][9][10].
The human face is integral for communication, person-
specific information, identity validation, and emotional
comprehension via facial cues. The distinctiveness of facial
features and their numerous parameters facilitate accurate
person identification. Human beings effortlessly recognize
faces, leveraging neural cells specialized in discerning
specific local features like edges, lines, angles, and motion.
Computers, however, employ algorithms to extract
distinctive facial characteristics such as skin tone, facial
orientation, shape. The objective of this project was real-
time detection and recognition of individuals using a
webcam. Various platforms exist for creating machine

vision applications, but this endeavor utilized Python,
coupled with OpenCV – an open-source software library
catering to real-time computer vision and image
processing. Python's compatibility with OpenCV makes it a
reliable choice for webcam-based real-time face detection
and recognition, extending to images. For the recognition
of the face the face detection algorithm is used. Haar
cascade algorithm is used in this work. Its an machine
learning object detection algorithm. To achieve good
accuracy the cascade function is trained using the negative
and positive images. The use of cascade is to do the feature
extraction by filters. These filters are Haar Features called
as Edge, line and rectangle features. The project
commenced with face detection using a pre-trained Haar
Cascade classifier for eyes and faces. Likewise, for
recognition, a classifier was trained on multiple facial
images, each associated with a unique identity. This trained
classifier was subsequently deployed for real-time
recognition. The simplicity and readability of OpenCV
contribute to its popularity for image and video processing,
compatible with a wide array of operating systems
This work is divided into three categories which begins
with detection of face by data gathering, training and
recognition. Figure 1 shows three steps to detect a face.
Fig 1:Steps for face recognition
3.1.1 Face Detection and Data Gathering
In the first step face detection and face gathering was done
by using OpenCV. Secondly Haar cascade model was
developed for training purpose. Sample faces have been
extracted and trained. Then the trained model is used for
recognition. Images are taken in the form of RGB channel.
OpenCV stores these images in Blue, Green and Red
channel. So the image detected is shown in BGR channel.
3.1.2 Training
A dataset as shown Figure. 2 has been taken for training.
To detect and identify particular person the dataset will be
trained with specific identification. Each training image
with similar pixel value will be taken for training.
Fig 2: Image datasets
3.1.3 Recognition
The trained dataset had been taken for recognition. All the
images have been extracted, cropped, resized, are being
converted to grayscale. A classifier had been trained to
classify and recognize the person. The algorithm was
trained using the dataset of facial images. Each image used
for recognition has the same ID, the classifier had been

trained using previously trained facial data is shown in
figure 3. Firstly, all the modules had been imported. The
classifier method had been created and the data directory
containing facial images had been taken in the classifier
method. Each image has a user ID and an image ID. Finally,
the system performs face recognition. The detected and
recognized face will be displayed with the name. The
recognizer labeled the face as varun, deeraj shivteja and
tasleem etc. Figure. 3 shows the final recognized faces the
person when given with different inputs. If any new
persons face comes into the camera, based on the facial
features. the system recognizes the face.
Fig 3: Recognized images
4. CONCLUSIONS
Artificial intelligence has propelled various sectors to
unprecedented heights. The integration of AI and machine
learning has revolutionized every aspect of society,
tackling seemingly impossible challenges. Machine
learning, as a subset of AI, empowers systems to
autonomously learn and improve through data analysis
and experience. These technologies have found
widespread application in industries like robotics,
healthcare, computer vision and gaming. The benefits of
technological advancements are manifold, significantly
elevating living standards in both developed and
developing countries. AI, essentially computer software
designed to simulate human intelligence, is capable of
processing various forms of data such as images, text,
video and audio. In this work, real-time face detection and
recognition were accomplished using Haar Cascade
machine learning object detection algorithm. This
algorithm identifies objects within an image, particularly
focusing on facial features like eyes and overall structure.
The process involves constructing part features, line
features, and four-rectangle features to pinpoint facial
images accurately. A cascade classifier aids in assessing
the accuracy of the detection, sliding through the image to
determine positive or negative matches. The pervasive use
of AI and machine learning has become increasingly
prevalent across various sectors, ushering in a new era of
human progress and improving overall living standards.
REFERENCES
[1] Farfinkle, S. 2002. iRobot Roomba. MIT Technology
Review. Available:
https://www.technologyreview.com/s/401687/irobo
t-roomba/. Accessed on Dec 14th, 2019.
[2] Flainski, M. 2016. Introduction to Artificial
Intelligence. Springer International Publishing.
Accessed on Dec 4th, 2019.
[3] Beklemysheva, A. 2020. Why Use Python for AI and
Machine Learning. Available:.
https://steelkiwi.com/blog/python-for-ai-and-
machine-learning/. Accessed on 10th March
2020Gouillart, E. 2020. Scikit-Image: Image
Processing. Available:
https://scipylectures.org/packages/scikit-
image/index.html. Accessed on March 25th, 2020.
[4] Helterman, R. 2011. Learning to Program with
Python. Southern Adventist University. Accessed on
March 13th, 2020.
[5] Isoni, A. 2016. Machine learning for the web.
Birmingham: Packet publishing Ltd. Accessed on
March 13th, 2020.

[6] Kulkarni, P. & Joshi, P. 2015. Artificial Intelligence.
Delhi: PHI Learning Private Limited. Accessed on Dec
21st, 2019
[7] Li, D. & Du, Y. 2017. Artificial intelligence with
uncertainty. Beijing, China: CRC Press. Accessed on
Dec 11th, 2020.
[8] Nicholson, C. 2019. AI vs ML vs DL. AI Wiki. Available:
https://pathmind.com/wiki/ai-vs-machinelearning-
vs-deep-learning. Accessed on Feb 26th, 2020.
[9] Russell, S. & Norvig, P. 2003. Artificial Intelligence A
Modern Approach. 2. New Jersey: The MIT Press.
Accessed on Dec 13th,:
https://courses.cs.washington.edu/courses/csep590/
06au/projects/history-ai.pdf. Accessed on 12th Dec
2019.
[10] Solem, J. 2012. Programming Computer Vision with
Python. Sebastopol: O'Reilly Media. Accessed on
March 16th, 2020.
BIOGRAPHIES
P.V.K.S Varun
B.Tech,Department of Electronics
and Computer Science
Engineering
Mahindra University,
Hyderabad
Syed.Tasleem
B.TECH(CSE)
Narayana Engineering College,
Gudur-524101
R. VenkataSivateja
B.TECH(CSE)
Narayana Engineering College,
Gudur
T.Dheeraj Reddy
B.Tech,Department of Computer
Science and Engineering,
Mahindra University,
Hyderabad
hor

Face Detection Using Artificial Intelligence and Machine Learning with Python

More Related Content

Similar to Face Detection Using Artificial Intelligence and Machine Learning with Python (20)

More from IRJET Journal (20)

Recently uploaded (20)

Face Detection Using Artificial Intelligence and Machine Learning with Python