Face Recognition Attendance System 1

Uploaded by

Parvathy

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

21 views19 pages

Face Recognition Attendance System 1

Uploaded by

Parvathy

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 19

College of Engineering Poonjar

FACE RECOGNITION ATTENDANCE SYSTEM

ALENSE K T
LIYA GIJO
POOJA KRISHNAN

December 28, 2024

Guided by : FLOWER ABRAHAM MUNDACKAL(Asst.Prof in ECE)

ALENSE K T LIYA GIJO POOJA KRISHNAN FACE RECOGNITION ATTENDANCE SYSTEM 1

Index
College of Engineering Poonjar

1 Problem Statement and Objective

2 Literature Review

3 Algorithm And Model

4 Block Diagram

5 Tools

6 Project Result

7 Conclusion

8 Reference

ALENSE K T LIYA GIJO POOJA KRISHNAN FACE RECOGNITION ATTENDANCE SYSTEM 2

Problem Statement and Objective
Problem Statement and Objective College of Engineering Poonjar

In most educational institutions the attendance is taken manually. It

is not only time consuming, but it is also unsecured and unreliable
and it can be lost.
OBJECTIVE
Get a less false-positive rate.
The attendance system is used to track and monitor whether a
student attends a class.
Eliminate duplicate data entry and errors in time and attendance
entries.
Eliminate paperwork and save time

ALENSE K T LIYA GIJO POOJA KRISHNAN FACE RECOGNITION ATTENDANCE SYSTEM 3

LITERATURE REVIEW
Literature Review College of Engineering Poonjar

EIGEN FACES SCALE INVARIANT FEATURE

• Appearance-based approach to • SIFT is used to extract distinc-
face recognition tive invariant features from im-
ages.
• Extract the relevant facial in- • It can also be used to perform
formation, which may or may not reliable matching.
be directly related to human intu-
ition of face features
• To reduce the computation and • For image matching and recog-
space complexity, each face image nition, SIFT features are extracted
can be represented using a small from a set of reference images and
number of parameters. stored in database.

ALENSE K T LIYA GIJO POOJA KRISHNAN FACE RECOGNITION ATTENDANCE SYSTEM 4

LITERATURE REVIEW
Literature Review College of Engineering Poonjar

LIMITATION
EIGEN FACES SCALE INVARIANT FEATURE
• Lack of discriminant power. • It is mathematically compli-
cated and computationally heavy.
• Applicable only to front views. • SIFT is based on the Histogram
of Gradients.So the gradients of
each Pixel in the patch need to
be computed and these computa-
tions cost time.
• Good performance only under • It is not effective for low pow-
controlled background(not includ- ered devices.
ing natural scenes).

ALENSE K T LIYA GIJO POOJA KRISHNAN FACE RECOGNITION ATTENDANCE SYSTEM 5

ALGORITHM AND MODEL
Algorithm And Model College of Engineering Poonjar

The Algorithm used is Haar Cascade Algorithm and Local Binary Pattern
Algorithm.
METHODOLOGY
Enrollment Stage
Collect images of each student
Face Detection- Haar cascade algorithm
Slide a fixed size window across our image at multiple scales
At each of these phases, our window stops, computes some features,
and then classifies the region as face or not.
The features considered are:Edge feature, line feature,Four rectangular
feature
Every single pixel being replaced by an arrow (gradients)
Break the image into small squares
In each square, count up how many gradients point in each major
direction
ALENSE K T LIYA GIJO POOJA KRISHNAN FACE RECOGNITION ATTENDANCE SYSTEM 6
ALGORITHM AND MODEL
Algorithm And Model College of Engineering Poonjar

Replace that square in the image with the arrow directions that were
strongest gradient
Face landmark Estimation and Encode Faces
128 measurements for each face
Assign user id for each student and tag it with corresponding faces
store the details in the database.
Recognition Stage
For attendance ,from the recorded video of the class we slice images of
all students
Face Detection-Haar Cascade Algorithm
Face Recognition-LBPH Algorithm
First, convert the image frame into a matrix form of 3X3 pixels
if the neighboring pixel in the matrix is larger than the center pixel of
that matrix the value will be given as 1 and 0 is given to lower pixel
area

ALENSE K T LIYA GIJO POOJA KRISHNAN FACE RECOGNITION ATTENDANCE SYSTEM 7

ALGORITHM AND MODEL
Algorithm And Model College of Engineering Poonjar

Convert the binary number to the decimal number and convert it to

the average pixel value of the matrix
Based on the image obtained, we can extract the histogram of each
region which represents the characteristics of the original image
To find the image that matches the input image we just need to
compare two histograms and return the image with the closest
histogram
We use Euclidien distance to calculate the distance between two
histograms
Find the person in our database of known people who has the closest
measurements to our test image
Attendance marking in the excel file

ALENSE K T LIYA GIJO POOJA KRISHNAN FACE RECOGNITION ATTENDANCE SYSTEM 8

BLOCK DIAGRAM
Block Diagram College of Engineering Poonjar

ALENSE K T LIYA GIJO POOJA KRISHNAN FACE RECOGNITION ATTENDANCE SYSTEM 9

TOOLS
Tools College of Engineering Poonjar

Programming Language-Python