Scribd
Upload
277 views5 pages

How To Write Thesis Proposal On Face Recognition

This is a thesis proposal for university final year students of CSE. This is a sample. Student will be able to write a new thesis proposal using this sample.

Uploaded by

Rahajul Amin Raju
Copyright
© © All Rights Reserved
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
277 views5 pages

How To Write Thesis Proposal On Face Recognition

This is a thesis proposal for university final year students of CSE. This is a sample. Student will be able to write a new thesis proposal using this sample.

Uploaded by

Rahajul Amin Raju
Copyright
© © All Rights Reserved
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/ 5

Face Recognition using Image Processing

Name: Md. Rahajul Amin


Id: 172311006
Section: A
Batch: 15th
Dept. of CSE
Varendra University

Introduction
Real time face recognition is part of the field of biometrics. Biometrics is the ability for a computer
to recognize a human through a unique physical trait. Face recognition provides the capability for
the computer to recognize a human by facial characteristics.
A facial recognition device is a device that takes an image or a video of a human face and compares
it to other image faces in a database. The structure, shape and proportions of the faces are
compared during the face recognition steps. In addition, distance between the eyes, nose, mouth
and jaw, upper outlines of the eye sockets, the sides of the mouth, location of the nose and eyes,
and the area surrounding the check bones are also compared [1].

Background and Context


Generally we know about face recognition. It is used for security purposes and for authentication. I
want to do thesis on face recognition using image processing technique so that I can recognise or
detect faces and there should be Artificial Intelligence (AI) technology and Machine Learning (ML)
technology for automatic recognition system. I want to do something which will be very smart
technology.

Problem Statement
Nowadays security system for home, office or other secure places is very important. So, if we make
an easy system to recognize people whom are not involved with that institution or home then the face
recognition system will identify unknown person and alert the authority.
This system will be also used for authentication known people. Suppose If will apply this system on
home security then this system will try to match faces of all persons whom are stored in database and
send success message to the authority. If it can’t match faces with database’s data then it will send
data unsuccessful message to the authority.

The main purpose of this system will be to identify known and unknown people by using a camera
and the system which I am going to make. It will be the main work of the face recognition technology.
Literature review

Many face recognition software have been implemented during the past decade. Each software
uses different methods and different algorithm than other software.

Key Concepts, Theories and Studies

 Some facial recognition software extracts the face features from the input image to identify
the face [1].
 Other algorithms normalize a set of face images and then compress the face data, the saves
the data in one image that can be used for facial recognition. The input image is compared
with the face data [2].
 New method for face recognition is being used which is the three-dimensional facial
recognition. In this method, a 3-D sensor is used to capture information about the shape of
the face so that only distinctive features of the face, such as the contour of eye sockets, nose
and chin, are used for face recognition [1].
 Another new technique in facial recognition is called skin texture analysis. This technique
uses the visual details of the skin, as captured in standard digital or scanned images and then
turns the unique lines, patterns, and spots apparent in a person’s skin into a mathematical
space.

Research design and methods


There are three main tasks of face recognition may be named: “document control”, “access
control”, and “database retrieval”.

Research design

It will be a fully automated system, user friendly, turnkey mobile and live-video surveillance facial
recognition system. This software generates an alert whenever a face is recognized; and this occurs
when the match of the input face with a face in the database is above a user defined probability.
The advantage of the system is the availability to work in low resolution environments enabling real-
world performance [1].
Below a flowchart which shows the basic features of the face recognition system.
Methods and Sources

First Algorithm for Face Detection:


First, the image is taken as an input into the software. The program then converts it from its color
mode (RGB) into grayscale then resizes it. An edge detection operation is then applied by calculating
the gradient of the image. To calculate the gradient of an image, the Sobel operator is used, which
creates a binary mask using a user-specified threshold value. The threshold value is determined by
getting the mean of all gray values in the image. The next step is a dilation operation in order to
make the borderlines thicker, followed by a filling technique in order to fill the hole in the face. After
holes filling, erosions followed by dilations are applied to get rid of other smaller objects in the
image and maintain the region of interest intact. Finally, the image is resized back to its original size.

1) Filtering
After taking the initial picture, the median filter technique (The median filter is a non-linear digital
filtering technique, often used to remove noise from an image or signal. Such
noise reduction is a typical pre-processing step to improve the results of later processing)
is applied on the image. The purpose of this technique is to eliminate the noise that will appear during
the capturing, and to enhance the edging procedure. So when using other kind of filter it will affect
the edging and an error will appear.
2) Resizing
After filtering the image, it is resized from 2048 x 1536 x 3 (resolution of the camera) to 512 x 384 x
3, because it easier to MATLAB to deal with small image and the processor will be faster.
3) Color Mode conversion and Skin Detection
The initial image taken from the camera is in RGB color mode which is not good to analyze, because
the luminous and the ceruminous are mixed together and then the removing of the important
parameter will be difficult, so the picture will be converted to (HSV) that facilitate the process of
taking the important parameters.
HSV: stand for Hue, Saturation and value and also it is called HSB for brightness [4].

4) Morphological Operations
After doing all previous procedure, the image will be processed using morphological operations. It
has 3 steps:
a) Dilation
b) Filling
c) Erosion

5) Elimination of the non Face Regions


After doing the previous functions, the elimination of the non-face area will be performed using two
techniques:
a) Pixels Technique
b) Ratio Technique

**Above steps are the basic steps of image processing for the first algorithm which is based on an
image.

Second Algorithm for Face Detection:


In the Recognition procedure, the initial image will be gray scaled to accelerate the processing. After
applying the face detection algorithm, only the face part is used now to be compared with the database
face images.
First, the face is detected using the face detection algorithm discussed previously. The face is split
into three sub-images emphasizing on the special features of the face. Then, a correlation is performed
on each sub-image with the database images followed by results averaging to check if the face is
recognized or not[5]. It has 4 steps:
a) Lightening Effect
b) Scaling
c) Correlation
d) Segmentation
Research schedule/Time frame

Research phase/stage Objectives/Goals Deadline


Research on first algorithmTry to implement first 10th semester(before final)
and image processing algorithm.
Research on SecondImplement second algorithm 11th semester(before
algorithm and about reads afor good accuracy and face midterm)
real-time video imagedetection from a real-time
processing video.
Research on Neural Network Implement more important 11th semester(before final)
and Trained Database things.
Calculate accuracy Output of my research 12th semester(before
midterm)
Research for more accuracy Try to complete full thesis 12th semester(before final)
on face detection

References

 The name of the paper I have read to prepare this template is Face Recognition Security
System

[1] FindBiometrics, Faceial recognition, [Online], Available at: http://findbiometrics.com/solutions/facial-recognition/.


[2] Steve Mann, “Intelligent Image Processing”, Wiley-Interscience 2002.
[3] Walter G.Kropatsh, “Digital Image Analysis”, Springer 2002.
[4] Wikipedia, HSL and HSV [Online], Available at: http://en.wikipedia.org/wiki/HSL_and_HSV.
[5] Bernd Jahne, “Digital Image Processing”, 5th Edition, Springer 2002.

You might also like