Department of ICT

A project(minor) report defense on:

“Face Recognition Attendance System”

Project Members: Submitted to:

Sandip Subedi 190148 Department of ICT
Sahil Poudel 190147 Cosmos College of
Devraj Chaudhary 190144 Management and Technology
Mahesh Bhatt 190111

Date : 28th Sep, 2023

1
Department of ICT, Cosmos College of Management and Technology
 OUTLINE
• Background​
• Objectives​
• Statement of the problem​
• Scope and limitations​
• Literature Review​
• Requirements Analysis
• Methodology​
• Results And Discussion​
• References
• Appendix

Department of ICT, Cosmos College of Management and Technology 2

 BACKGROUND
• The traditional method of taking attendance in classrooms, offices, and
other institutions involves using paper and pen, which can be tedious,
time-consuming, and prone to errors.
• The Face Recognition Attendance System (FRAS) is one such system that
uses facial recognition technology to record the attendance of individuals.
• This proposal aims to develop a Face Recognition Attendance System that
will be used in educational institutions and workplaces to automate the
attendance process.
• The system will also be customizable to meet the unique requirements of
different organizations and will integrate with existing infrastructure
seamlessly.
• In this proposal, we will discuss the features, benefits, and
implementation plan of the proposed FRAS.

Department of ICT, Cosmos College of Management and Technology 3

 STATEMENT OF PROBLEM
• The traditional method of taking attendance using paper and
pen is time-consuming, prone to errors, and can be easily
manipulated.
• Manual attendance records also require teachers or
administrators to spend a considerable amount of time
maintaining them.
• Contactless attendance systems are needed to ensure the
safety and well-being of students and employees.
• Face recognition technology can accurately identify
individuals based on their facial features and does not require
physical contact.

Department of ICT, Cosmos College of Management and Technology 4

 OBJECTIVES
• To detect the face segment from the video frame.
• To extract the useful features from the face detected.
• To classify the features in order to recognize the face
detected.
• To record the attendance of the identified student.

Department of ICT, Cosmos College of Management and Technology 5

 SCOPE AND LIMITATIONS
Scope:
• Accurate Attendance Tracking: It eliminates the need for
manual attendance registers or swipe cards.
• Time Efficiency: The system can quickly process and match
faces against a database, enabling swift and efficient
attendance management.
• Accessibility: The system can be accessed remotely,
allowing users to mark their attendance from various
locations.
• Real-time Monitoring: It enhances transparency and
enables timely intervention, if required.
Department of ICT, Cosmos College of Management and Technology 6
 SCOPE AND LIMITATIONS
Limitations:
• Accuracy Dependency on Image Quality, Factors such
as lighting conditions, occlusions, and pose variations
can affect the accuracy of the system.
• Vulnerability to Spoofing
• Hardware Requirements
• Privacy Concerns, Implementing a face recognition
attendance system raises privacy concerns as it
involves capturing and processing biometric data.
• Cost Considerations
Department of ICT, Cosmos College of Management and Technology 7
 LITERATURE REVIEW
Attendance System:
• Arun Katara et al. (2017) mentioned disadvantages of RFID (Radio Frequency Identification) card
system, fingerprint system and iris recognition system.
• RFID card system is implemented due to its simplicity.
• However, the user tends to help their friends to check in as long as they have their friend’s ID
card.
• The fingerprint system is indeed effective but not efficient because it takes time for the
verification process so the user has to line up and perform the verification one by one.
• However for face recognition, the human face is always exposed and contain less information
compared to iris.
• Voice recognition is available, but it is less accurate compared to other methods. Hence, face
recognition system is suggested to be implemented in the attendance system.

Department of ICT, Cosmos College of Management and Technology 8

 LITERATURE REVIEW
Haar Cascade Detection
• Haar Cascade Detection is one of the oldest yet powerful
face detection algorithms invented.
• It has been there since long, long before Deep Learning
became famous.
• Haar Features were not only used to detect faces, but
also for eyes, lips, license number plates etc.
• The models are stored on GitHub, and we can access
them with OpenCV methods.

Department of ICT, Cosmos College of Management and Technology 9

 LITERATURE REVIEW
Local Binary Pattern Histogram
Local Binary Pattern (LBP) is a simple yet very efficient
texture operator which labels the pixels of an image by
thresholding the neighborhood of each pixel and consi-
ders the result as a binary number.
LBPH algorithm working steps:
1. Parameters
2. Training the Algorithm
3. Applying the LBP operation
4. Extracting the Histograms
5. Performing the face recognition
Department of ICT, Cosmos College of Management and Technology 10
 SOFTWARE AND HARDWARE
REQUIREMENT
Software Requirements:
1. Python
2. Sublime text Editor/py-charm/vs-code
3. OpenCV: We used OpenCV dependency for python . OpenCV is library
where there are lots of image processing functions are available.
• Derivation: Gradient/Laplacian computing, contours delimitation
• Hough transforms: lines, segments, circles, and geometrical shapes
detection
• Histograms: computing, equalization, and object localization with back
projection algorithm
• Segmentation: thresholding,distance transform,foreground/background
detection, watershed segmentation

Department of ICT, Cosmos College of Management and Technology 11

 SOFTWARE AND
HARDWARE REQUIREMENTS
• Filtering: linear and nonlinear filters, morphological operations
• Cascade detectors: detection of face, eye, car plates
• Interest points: detection and matching
• Video processing: optical flow, background subtraction, camshaft (object tracking)
• Photography: panoramas realization, high definition imaging (HDR), image
inpainting
Hardware Requirements:
• Windows 7 or higher
• I3 processor system or higher
• 4 GB RAM or higher
• 100 GB ROM or higher

Department of ICT, Cosmos College of Management and Technology 12

 METHODOLOGY
4.1 Software Process Model:
We have decided to use Incremental model for our project as it makes easier to
divide the work.

Department of ICT, Cosmos College of Management and Technology 13

 METHODOLOGY
Block Diagram
/Flowchart

Department of ICT, Cosmos College of Management and Technology 14

 METHODOLOGY
• Use Case Diagram

Department of ICT, Cosmos College of Management and Technology 15

 METHODOLOGY

ER Diagram:

Department of ICT, Cosmos College of Management and Technology 16

 METHODOLOGY

Activity Diagram

Department of ICT, Cosmos College of Management and Technology 17

 RESULTS AND DISCUSSION

EXPERIMENTAL RESULTS:
Error occurs in testing 10 student image
• Illumination Variations
• Pose Variations
• Occlusions and Partial Face Coverage
• Similar Facial Features
• Limited Training Data

Department of ICT, Cosmos College of Management and Technology 18

 RESULTS AND DISCUSSION contd..

Taking 50 samples as a training dataset

The number of samples in the training dataset should
be determined based on the following considerations:
• Limited Training Data
• Overfitting and Generalization
• Data Quality
In practice, the number of samples in the training dataset for LBP
can vary widely. Some studies may use a few hundred samples,
while others may use thousands or more.

Department of ICT, Cosmos College of Management and Technology 19

 RESULTS AND DISCUSSION contd..
Experimental result:
We performed a set of experiments to demonstrate the efficiency of the
proposed method.
50 different images of 10 persons are used in training set.

Department of ICT, Cosmos College of Management and Technology 20

 LIMITATIONS
• Our project is only applicable for certain number of
peoples.
• Camera quality can affect the accuracy for
recognizing.
• To detect a face, the candidate must be in an area
with light; otherwise, it will not be able to detect it.

Department of ICT, Cosmos College of Management and Technology 21

 FUTURE ENHANCEMENTS
• This project can be upgraded to many number of
peoples.
• Instead of taking the attendance of single person we
can extend the system to take the attendance by cctv
when the employee enters the office.
• Can build in fully web based system.
• Can use neural network for high accuracy.

Department of ICT, Cosmos College of Management and Technology 22

 CONCLUSION
• The purpose of the project attendance monitoring with
face recognition is to minimize the errors that occur in
the traditional attendance taking system.
• This project is mainly built using machine learning which
exhibits robustness in face recognition of the person with
greater accuracy.
• Current work is focused on the face detection algorithms
from images or video frames.
• The aim is to automate and make a system that is
useful to the organization such as an institute.

Department of ICT, Cosmos College of Management and Technology 23

 TIME SCHEDULES

Department of ICT, Cosmos College of Management and Technology 24

 REFERENCES
[1]. A brief history of Facial Recognition, NEC, New Zealand,26 May 2020.
[Online].Available:https://www.nec.co.nz/marketleadership
/publications-media/a-brief-history-offacialrecognition/

[2] A. Geitgey, "Machine Learning is Fun! Part 4: Modern Face Recognition

withDeepLearning".[Online].Avilablehttps://medium.com/@ageitgey/mac
hine-learning-is-funpart-4-modern-face-recognition-with-deep-learning-
c3cffc121d78 [Accessed:13-May2019]
[3]. https://www.questjournals.org/jses/papers/Vol5-issue-2/D05021829.pdf

Department of ICT, Cosmos College of Management and Technology 25

 SNAPSHOTS
Home Page

Department of ICT, Cosmos College of Management and Technology 26

 SNAPSHOTS
Registering Face

Department of ICT, Cosmos College of Management and Technology 27

 SNAPSHOTS
Training Face

Department of ICT, Cosmos College of Management and Technology 28

 SNAPSHOTS
Taking Attendance

Department of ICT, Cosmos College of Management and Technology 29

 Thank You!

Department of ICT, Cosmos College of Management and Technology 30