Face Recognition Software

SRM Institution of Science and Technology

Minor Project (18CSP107L)

November 10th, 2022

Submitted to:

Niharika Saxena

Submitted By:

Atulit Anand (RA1911026030043)

Aryan Tomar (RA1911026030031)

Pahuni Shrivastava (RA1911026030001)

Uttarak Som (RA1911026030009)

(Under Section 3 of UGC Act, 1956)

BONAFIDE CERTIFICATE

Certified that this project report titled “Face Recognition Software” is the bonafide work of “

Atulit Anand (RA1911026030043), Aryan Tomar (RA1911026030031), Uttarak Som

(RA1911026030009), Pahuni Shrivastava (RA1911026030001) ”, who carried out the project

work under my supervision. Certified further, that to the best of my knowledge the work reported

herein does not form any other project report or dissertation on the basis of which a degree or

award was conferred on an earlier occasion on this or any other candidate.

I would like to express my deepest gratitude to my guide, Dr. Niharika Saxena for her valuable

guidance, consistent encouragement, personal caring, timely help and providing me with an

excellent atmosphere for doing research. All through the work, in spite of her busy schedule, she

has extended cheerful and cordial support to me for completing this research work.

I would like to extend my gratitude towards DR. R.P. MAHAPATRA (Head Of The Department

- Computer Science and Engineering) as he always ensured that I get a practical overview of the

project. This project is an outcome of that exposure.

Uttarak Som, Atulit Anand, Aryan Tomar, Pahuni Shrivastava .

● Abstract

● Introduction ( Purpose, Scope and Method)

● Facial Recognition Software in Python

● Procedure

○ Face detection

○ Image Compression

○ Internal Assessment

○ Output Representation

● Results

● References
 Abstract

Face Recognition is an application of machine learning that is capable of detecting, tracking,

identifying and verifying human faces from an image (singular) or stream of images called a

video captured using a digital camera or any digital equipment. A lot of progress has been made

in the domain of face detection and recognition for security, identification and attendance

purposes yet there are still issues limiting the progress to surpass and reach human level real

world accuracy. ​

We addressed some of the issues hampering face recognition accuracy so as to improve the

overall accuracy of our software. Our experimentation results show that this is not only very

accurate and reliable but is also very robust in implementing real world application of face

recognition systems.
 Introduction

Purpose: This purpose of this report is to present a follow up on the 7th semester project by

producing a workable face recognition software.

Scope: This system can be used anywhere where facial recognition is required which may

include applications detecting items, objects and people, making interactive guides for users,

arranging objects, etc.

Method: Our team has built this system from scratch using python programming language and

using several libraries like face_recognition, cv2, simple_facrec, etc.

Face recognition often involves various steps in producing an acceptable output. Hence we target

this complex problem by breaking it into several sub problems like facial detection, facial

analysis, converting and manipulating data and face imprints.

Python is a high level programming language which aims at producing readable and super

efficient code. It is cherished by the data science community because of its simplicity and the

amount of third party and in-libraries methods it provides.

For solving the problem of face recognition we are using python as our primary coding language.
 Procedure

Face detection

For image detection we are using cv2 and simple_facerec to locate and load images

using a camera.

Image Compression

Now we are using the loaded images and converting them into an optimal form which our

modal can feasibly understand. We have first used the library cv2 to read images from our

database followed by compressing and encoding those images using a python library called

face_recognition.

Internal assessment and image processing

This is the core step of our program which aims at loading images from the database and

then comparing the image provided to us by the camera using different methods from the user

defined SimpleFacerec class. This class encapsulated all the api calls and different methods to

provide a cleaner, more optimal interface.

Output Representation

After comparing both the streams of images in a real time computer the identify the

person currently facing the camera if he/she matches the one in the database. If all conditions are

met we can see a box around that person’s face on the screen
 Results

As a result of out experiment we are delivering a working machine learning model that

can detect if a person facing the camera belongs to the set of images in its database. This can be

used in security systems , for attendance monitoring , for object detection, for smart user guides

etc.
 References

● Nicholas Furl, Lúcia Garrido, Raymond J. Dolan, Jon Driver, Bradley Duchaine;

Fusiform Gyrus Face Selectivity Relates to Individual Differences in Facial

Recognition Ability. J Cogn Neurosci 2011; 23 (7): 1723–1740. doi:

https://doi.org/10.1162/jocn.2010.21545

● Google Face Detection using NLTK Face detection | ML Kit | Google Developers

https://developers.google.com/ml-kit/vision/face-detection