Major Project F
Major Project F
A
Project Report
Submitted in partial fulfillment of the requirement for the award of degree of
Bachelor of Engineering
In
Computer Science & Engineering
Submitted to
RAJIV GANDHI PROUDYOGIKI VISHWAVIDYALAYA,
BHOPAL (M.P.)
Guided by Submitted By
Prof. Nand Kishore Sharma Madiha Khan 0875CS131066
Manali Rode 0875CS131067
Nabila Hashmi 0875CS131079
Page | 1
Declaration
I hereby declared that the work, which is being presented in the project entitled “Face
Recognition System” partial fulfillment of the requirement for the award of the degree of
Bachelor of Engineering in Computer Science & Engineering, submitted in the department of
Computer Science & Engineering at Acropolis Technical Campus, Indore is an authentic
record of my own work carried under the supervision of “Prof. Nand Kishore Sharma”. I have
not submitted the matter embodied in this report for award of any other degree.
Prof. Nand
Kishore Sharma
Page | 2
Acropolis Technical Campus
Department of Computer Science & Engineering
Certificate
The project work entitled “Face Recognition System” submitted by Madiha Khan
(0875CS131066), Manali Rode (0875CS131067) , Nabila Hashmi (0875CS131079) is
approved as partial fulfillment for the award of the degree of Bachelor of Engineering in
Computer Science & Engineering by Rajiv Gandhi Proudyogiki Vishwavidyalaya, Bhopal
(M.P.).
Page | 3
Acknowledgement
We are thankful to our institute Acropolis Technical Campus, Indore for providing us the opportunity
to apply and convert our theoretical knowledge into practical skills through this project work.
Such work requires inputs, efforts and encouragement from all walks of life. We are indeed fortunate
to get active and kind co-operation from many corners without which this endeavor wouldn’t have
been a success.
We express our immense gratitude to our project supervisor Prof. Nand Kishore Sharma for his
valuable guidance and support and his constantly coordinating with us that made us to complete this
project to meet our objectives.
A special vote of thanks to our Head of the Department, Prof. Prashant Lakhadwala for his
constant vigilance, support and follow up on this project which made us to work on our toes during
the course of this project.
We also acknowledge the kind support rendered by Director Dr. S.C Sharma towards conducting the
experimental studies and compilation of the project work.
Our deep sense of gratitude to each and every person involved directly or indirectly in this project
work with special thanks to our classmates.
Abstract
Human face detection and recognition by computer systems has become a major field
of interest. Face detection algorithms are used in a wide range of applications, such as
security control, video retrieving, biometric signal processing, human computer
Page | 4
interface, face recognitions and image database management. However, it is difficult to
develop a complete robust face detector due to various light conditions, face sizes, face
orientations, background and skin colors. In this report, we propose a face detection
method for color images. Our method detects skin regions over the entire image, and
then generates face candidates based on a connected component analysis.
Two of the most important aspects in the general research framework of face
recognition by computer are addressed here: face and facial feature detection, and face
recognition or rather face comparison. The best reported results of the mug-shot face
recognition problem are obtained with elastic matching using jets.
In this approach, the overall face detection, facial feature localization, and face
comparison is carried out in a single step. This report describes our research progress
towards a different approach for face recognition. On the one hand, we describe a
visual learning technique and its application to face detection in complex background ,
and accurate facial feature detection . On the other hand , a fast algorithm for 2D-
template matching is presented as well as its application to face recognition . Finally,
we report an automatic , real-time face recognition system.
Page | 5
Table of Content
Chapter 1: Introduction
Face detection locates and segments face regions in cluttered images. It has numerous
applications in areas like surveillance and security control systems, content-based image
retrieval, video conferencing and intelligent human computer interfaces. Some of the current
face-recognition systems assume that faces are isolated in a scene. We do not make that
assumption. The system segments faces in cluttered images.
With a portable system, we can ask the user to pose for the face identification task. This can
simplify the face-detection algorithm. In addition to creating a more cooperative target, we can
interact with the system in order to improve and monitor its detection. The task of face detection
is seemingly trivial for the human brain, yet it remains a challenging and difficult problem to
enable a computer /mobile phone/PDA to do face detection. The human face changes with
respect to internal factors like facial expression, beard, mustache, glasses, etc. is sensitive to
external factors like scale, lightning conditions, and contrast between face, background and
orientation of face. Thus, face detection remains an open problem. Many researchers have
proposed different methods for addressing the problem of face detection. Face detection is
classified into feature-based and image-basedtechniques.The feature-based techniques use edge
information, skin color, motion, symmetry, feature analysis, snakes, deformable templates and
point distribution. Image-based techniques include neural networks, linear subspace methods,
like eigen faces fisher faces etc.
The problem of face detection in still images is more challenging and difficult when compared to
the problem of face detection in video, since motion information can lead to probable regions
where faces could be located.
Page | 6
1.1 Motivation
Face detection plays an important role in today’s world. It has many real-world
applications like human/computer interface, surveillance, authentication and video
indexing. However research in this field is still young. Face recognition depends heavily
on the particular choice of features used by the classifier. One usually starts with a given
set of features and then attempts to derive a optimal subset (under some criteria) of
features leading to high classification performance with the expectation that similar
performance can also be displayed on future trials using novel (unseen) test data.
Page | 7
A. Initialization process
B. Recognition process
The Initialization process involves the following operations:
Acquire the initial set of face images called as training set.
Calculate the eigen-faces from the training set, keeping only the highest eigenvalues.
These M images define the face space. As new faces are experienced, the eigen-faces can
be updated or recalculated.
Calculate distribution in this M-dimensional space for each known person by projecting
his or her face images onto this face-space.
These operations can be performed from time to time whenever there is a free excess operational
capacity. This data can be cached which can be used in the further steps eliminating the overhead
of re-initializing, decreasing execution time thereby increasing the performance of the entire
system.
Having initialized the system, the next process involves the steps:
Calculate a set of weights based on the input image and the M eigen-faces by projecting
the input image onto each of the eigen-faces.
Determine if the image is a face at all (known or unknown) by checking to see if the
image is sufficiently close to a free space.
If it is a face, then classify the weight pattern as either a known person or as unknown.
Update the eigen-faces or weights as either a known or unknown, if the same unknown
person face is seen several times then calculate the characteristic weight pattern and
incorporate into known faces. The last step is not usually a requirement of every system
and hence the steps are left optional and can be implemented as when the there is a
requirement.
Page | 8
1.2 Thesis Goals:
The goal of the project is to put in place new solutions for Face Recognition. As per the end user
expectations so that the accuracy, speedup, digitalization of the system will be enhanced. To do
this:
(a) Provide an efficient system for security purpose.
(b) Provide effective solutions for accurate Face Identification and access
control
It is have to be more specific term, to focus about the main idea and the details to applicant facial
recognition in our project, that easier to all users, most quickly, and modern application.
(a) To discuss and summarize the process of facial recognition.
(b) To look at currently available facial recognition techniques..
(c) To design and develop a robust facial recognition algorithm, so the algorithm should be
usable in a simple and easily adaptable setup. This implies a single camera setup, preferably a
webcam, and no use of specialized equipment.
Face recognition has a number of strengths to recommend it over other biometric modalities in
certain circumstances, and corresponding weaknesses that make it an inappropriate choice of
biometric for other applications. Face recognition as a biometric derives a number of advantages
from being the primary biometric that humans use to recognize one another. Some of the earliest
identification tokens, i.e. portraits, use this biometric as an authentication pattern. Furthermore it
is well-accepted and easily understood by people, and it is easy for a human operator to arbitrate
Page | 9
Machine decisions — in fact face images are often used as a human-verifiable Backup to
automated fingerprint recognition systems
The report is divided into four chapters. Each chapter deals with the different aspects of Face
Recognition Each chapter has various respective points explaining in detail:
Chapter 1: Introduction
Chapter 1 discusses about the basic introduction along with motivation to choose this project. It
includes some of the solution strategy, thesis goals and objective. Short Explanation of the
proposed work that is going to be implemented as the future work.
Page | 10
Chapter 2: General Introduction
This chapter contains the various terminology used in Face Recognition to have clear understanding
about the subject.
Principal Component Analysis, PCA, is a very widely used technique for dimensionality
reduction. The objective of PCA is to transform the representation space X into a new space Y, in
which the data are uncorrelated. The covariance matrix in this space is diagonal. The PCA
method leads to find the new set of orthogonal axis to maximize the variance of the data. The
final objective is dimensionality reduction of the feature vector. The steps to compute the PCA
are the following.
Step 1. The covariance matrix CovX is calculated over the input vectors set xi that corresponds
to i facial images represented as vectors x . The covariance is defined as ( )( ) T n i i i n ∑= − − −
= 1 1 1 CovX x x x x (6) where x denotes the mean of each variable of the vector x , and n is the
number of input vectors.
Step 2. The n eigenvalues of CovX are extracted and defined as 1 2 , ,... λ λ λn , where 1 2 ... λ ≥
≥≥ λ λn .
Step 3. The n eigenvectors are 1 2 , ,... ΦΦ Φn and are associated to 1 2 , ,... λ λ λn .
Step 4. A transformation matrix, WPCA , is created 1 2 [ , ,... ] WPCA n =Φ Φ Φ .
Step 5. The new vectors Y are calculated using the following equation
Page | 11
multiple of v. This condition can be written as the equation where λ is a scalar in the field F,
known as the eigenvalue, characteristic value, or characteristic root associated with the
eigenvector v. In the eigenspace, all the eigenvectors are not equally informative.
The differences between the training and testing feature vectors of each segment of the known
and unknown faces are used to compute the segmental Euclidean distance that serves as the
classifier.
All the M by N images are easily discussed in an MN dimensional Euclidean space, called image
space. It is natural to adopt the base e1, e2, · · · , eMN to form a coordinate system of the image
space, where ekN+l corresponds to an ideal point source with unit intensity at location (k, l).
Thus an image x = (x 1 , x2 , · · · , xMN ), where x kN+l is the gray level at the (k, l)th pixel, is
represented as a point in the image space, and x kN+l is the coordinate with respect to ekN+l .
The origin of the image space is an image whose gray levels are zero everywhere.the traditional
Euclidean distance is sensitive to deformation and translation due to the lack of consideration of
the spatial relationship of pixels. IMED, to a certain extent, overcomes this defect
The attractiveness of neural networks lies in the easiness with which non linearity can be
introduced in the network model. The earliest artificial neural network (ANN) used for the face
recognition is a single layer adaptive network called WISARD and such a separate network is
required to store the model of a person11. Other networks for the face detection include
multilayer perceptron12 and convolution neural network13. A multi- resolution pyramid
structure is employed for the face recognition. A hybrid neural network combines the local image
sampling, a self-organizing map (SOM) neural network and a convolution neural network
2.5 2DPCA
2DPCA is based on 2D image matrices rather than 1D vectors so the image matrix does not need
to be transformed into a vector prior to feature extraction. Instead, an image covariance matrix is
Page | 12
constructed directly using the original image matrices, and its eigenvectors are derived for image
feature extraction
Let X denote an n-dimensional unitary column vector. Our idea is to project image A, an m n
random matrix, onto X by the following linear transformation
: Y ¼ AX: ð1Þ
Thus, we obtain an m-dimensional projected vector Y, which is called the projected feature
vector of image A. How do we determine a good projection vector X? In fact, the total scatter of
the projected samples can be introduced to measure the discriminatory power of the projection
vector X. The total scatter of the projected samples can be characterized by the trace of the
covariance matrix of the projected feature vectors. From this point of view, we adopt the
following criterion: J
ðXÞ ¼ trðSxÞ;
Page | 13
matrices L = [u1, ··· , u1 ] ∈ IRr×1 and R = [v1, ··· , v2 ] ∈ IRc×2 . Then the projection of X ∈
LT XR ∈ R1×2 .
2DLDA, is presented for dimension reduction. 2DLDA is an extension of LDA. The key
difference between 2DLDA and LDA is that 2DLDA works on the matrix representation of
images directly, while LDA uses a vector representation. 2DLDA has asymptotically minimum
memory requirements, and lower time complexity than LDA, which is desirable for large face
datasets, while it implicitly avoids the singularity problem encountered in classical LDA.
2.8 Fuzzy Logic
Fuzzy Rule Based system which is used for facial expression recognition from facial features.
Fuzzy logic can be used to form linguistic models and comes with a solid qualitative base. Fuzzy
systems have been used in many classification and control problems including facial expression
recognition.
The Algorithm of Fuzzy 2DPCA
The key step of fuzzy 2DPCA is how to incorporate the contribution of each training sample into
the total scatter matrix. Based on the fuzzy set theory, each sample can be classified into
multiclasses with fuzzy membership degrees, instead of binary classification, so, in the
redefinition of the fuzzy total scatter matrix, the membership degree of each sample (contribution
to each class) and the class information should be considered. The idea of the fuzzy 2DPCA is
that the means of each class are calculated with fuzzy membership degrees matrix of all training
samples firstly, and then we average all class means to get the center of all training samples. In
the definition of the new proposed method, the original mean of all samples is replaced with the
new obtained fuzzy mean . Based on what we have described above, the algorithm of the
proposed supervised fuzzy 2DPCA can be summarized as follows.
Page | 14
Step 1 (FKNN). The fuzzy membership degree matrix can be computed with the FKNN
algorithm in the original training image space.
Step 2. Based on , compute the fuzzy class mean of each class; after that, the average procedure
is performed on all fuzzy class means to get the fuzzy average of all training samples
Step 3. Redefine the total scatter matrix according to the new obtained fuzzy average vector .
Then the optimal projection matrix can be obtained by computing the optimal problem.
Step 4 Project all samples into the obtained optimal discriminant matrix and classify testing
samples with nearest distance classifier.
Feature extraction starts from an initial set of measured data and builds derived values (features)
intended to be informative and non-redundant, facilitating the subsequent learning and
generalization steps, and in some cases leading to better human interpretations. Feature
extraction is related to dimensionality reduction.
When the input data to an algorithm is too large to be processed and it is suspected to be
redundant (e.g. The same measurement in both feet and meters, or the repetitiveness of images
presented as pixels), then it can be transformed into a reduced set of features (also named
a features vector). This process is called feature selection. The selected features are expected to
contain the relevant information from the input data, so that the desired task can be performed by
using this reduced representation instead of the complete initial data.
Feature extraction involves reducing the amount of resources required to describe a large set of
data. When performing analysis of complex data one of the major problems stems from the
number of variables involved. Analysis with a large number of variables generally requires a
large amount of memory and computation power, also it may cause a classification algorithm
to overfit to training samples and generalize poorly to new samples. Feature extraction is a
general term for methods of constructing combinations of the variables to get around these
problems while still describing the data with sufficient accuracy.
Page | 15
Many data analysis software packages provide for feature extraction and dimension reduction.
Common numerical programming environments such as MATLAB, SciLab, NumPyand the R
language provide some of the simpler feature extraction techniques (e.g. principal component
analysis) via built-in commands. More specific algorithms are often available as publicly
available scripts or third-party add-ons.
The Backpropogation algorithm is a sensible approach for dividing the contribution for each
weight. The algorithm can be decomposed in the following four steps: i) Feed-forward
computation ii) Back propagation to the output layer iii) Back propagation to the hidden layer iv)
Weight updates The algorithm is stopped when the value of the error function has become
sufficiently small. This is very rough and basic formula for BP algorithm
2.10.1 Feed-forward computation
Feed forward computation or forward pass is two step process. First part is getting the values of
the hidden layer nodes and second part is using those values from hidden layer to compute value
or values of output layer. Input values of nodes N0,0 and N0,1 are pushed up to the network
towards nodes in hidden layer ( N1,0 and N1,1)
2.10.2 Back propagation to the output layer
Next step is to calculate error of N2,0 node. From the table in figure 4, output should be 1.
Predicted value (N2,0) in our example is 0.464381. Error calculation is done the following way:
N2, 0Error = n2, 0∗(1−n2, 0)∗(N2, 0Desired−N2, 0) = 0.464381(1−0.464381)∗(1−0.464381) =
0.133225 Once error is known, it will be used for backward propagation and weights adjustment.
It is two step process. Error is propagated from output layer to the hidden layer first. This is
where learning rate and momentum are brought to equation. So weights W1,0 and W1,1 will be
updated first. Before weights can be updated, rate of change needs to be found. This is done by
multiplication of the learning rate, error value and node N1,0 value.
2.10.3 Back propagation to the hidden layer
Page | 16
Now errors has to be propagated from hidden layer down to the input layer. This is bit more
complicated than propagating error from output to hidden layer. In previous case, output from
node N2,0 was known beforehand. Output of nodes N1,0 and N1,1 was unknown. Let’s start
with finding N1,0 error first. This will be calculated multiplying new weight W1,0 value with
error for the node N2,0 value. Same way error for N1,1 node will be found
Page | 17
considerably with the advent of computerized databases and the digitization of analog data,
allowing for almost instantaneous personal identification.
2.12 Image restoration
Image Restoration is the operation of taking a corrupt/noisy image and estimating the clean, original
image. Corruption may come in many forms such as motion blur, noise and camera mis-focus.[1] Image
restoration is performed by reversing the process that blurred the image and such is performed by imaging
a point source and use the point source image, which is called the Point Spread Function (PSF) to restore
the image information lost to the blurring process.
Image restoration is different from image enhancement in that the latter is designed to emphasize features
of the image that make the image more pleasing to the observer, but not necessarily to produce realistic
data from a scientific point of view. Image enhancement techniques (like contrast stretching or de-
blurring by a nearest neighbor procedure) provided by imaging packages use no a priori model of the
process that created the image.
With image enhancement noise can effectively be removed by sacrificing some resolution, but this is not
acceptable in many applications. In a fluorescence microscope, resolution in the z-direction is bad as it is.
More advanced image processing techniques must be applied to recover the object .
In imaging science, Image Processing is processing of images using mathematical operations by using
any form of signal processing for which the input is an image, a series of images, or a video, such as
a photograph or video frame; the output of image processing may be either an image or a set of
characteristics or parameters related to the image. Most image-processing techniques involve treating the
image as a two-dimensional signal and applying standard signal-processing techniques to it. Images are
also processed as three-dimensional signals where the third-dimension being time or the z-axis.
Image processing usually refers to digital image processing, but optical and analog image processing also
are possible. This article is about general techniques that apply to all of t hem. The acquisition of images
(producing the input image in the first place) is referred to as imaging.
Page | 18
2.14 Template matching
Template matching is a technique in digital image processing for finding small parts of an image
which match a template image. It can be used in manufacturing as a part of quality control, a way
to navigate a mobile robot or as a way to detect edges in images. If the template image has strong
features, a feature-based approach may be considered; the approach may prove further useful if
the match in the search image might be transformed in some fashion.
For templates without strong features, or for when the bulk of the template image constitutes the
matching image, a template-based approach may be effective. As aforementioned, since
template-based template matching may potentially require sampling of a large number of points,
it is possible to reduce the number of sampling points by reducing the resolution of the search
and template images by the same factor and performing the operation on the resultant downsized
images, providing a search window of data points within the search image so that the template
does not have to search every viable data point, or a combination of both.
Page | 19
Chapter 3: Literature Reviews
3.1 Literature Review
4) Work Description :
In the beginning face recognition was treated as a 2Dpattern recognition problem. The
distances between important points were used to recognize known faces, e.g. measuring
the distance between the eyes or other important points or measuring different angles of
facial components. But it is necessary that the face recognition systems to be fully
automatic.
Face recognition is such a challenging yet interesting problem that it has attracted
researchers who have different backgrounds: psychology, pattern recognition, neural
networks, computer vision, and computer graphics. The following Three methods are
used to face recognition.
3. Hybrid Methods
5) Advantages : It is also clear that the recognition rate increases with the number of
training images per person. It is obvious that if the minimum distance between the
test image and other images is zero, the test image entirely matches the image from
the training base.
Page | 20
6) Limitations: If the distance is greater than zero but less than a certain threshold, it is
a known person with other facial expression, otherwise it is an unknown person
Literature Review 2
1) Paper Author : K Liton Chandra Paul, Abdulla Al Sumam
4) Work Description :
5) Advantage :
Page | 21
The advantage of this approach over other face recognition systems is in its simplicity.
Speed and insensitivity to small or gradual changes on the face.
6) Limitations:
This Proposed Algorithm was limited to Vertical Front View .
It worked only on still images and no real time face Recognition could be made possible.
Problem of huge database could not be overcome through this proposed algorithm.
It could only recognize images of fixed size.
Literature Review 3
4) Work Description :
The commonly used face recognition methods can be divided into several categories the
method based on geometric features, the method based on model the method based on the
statistical, the method based on neural network method and the method of combining
multiple classifiers, this paper mainly focuses on the method based on statistics. The
method based on statistical faced the image as random vector, thus using some statistical
methods to analyze the face model, the most representative methods include the
eigenface method based on principal component analysis .
5) Advantages :
This method retains the advantages of the original method at the same time, and used the
integrated advantages to make up for the deficiency of the original method, so this
method will achieve a good recognition effect
Page | 22
.In addition, this method combines the use of low frequency subspace and high frequency
subspace, which can not only save time, but also can ensure the rate of recognition.
Robust To Illumination Changes
6) Limitations:
The PCA and LDA methods have a common disadvantage, the two methods both need
deal with vector data, and they can not directly deal with image data, so before using
these two methods it need transform the image data to vector data firstly, but the
dimension of the processed data often is very high.
In addition, the two methods both need decompose the matrix feature value, but
computation for high dimensional data is very difficult and time-consuming
Literature Review 4
1) Paper Author :
Prof.Ujval Chaudhary, Chakoli Mateen Mubarak, Abdul Rehman,
Ansari Riyaz, Shaikh Mazhar
2) Journal : International Journal of Modern Engineering Research
3) Year : Volume 2, Issue.3 ,May-June 2012
4) Work Description :
Principal Component Analysis:
To do this, a face image is projected to several face templates called eigenfaces which can be
considered as a set of features that characterize the variation between face images. Once a set of
eigenfaces is computed, a face image can be approximately reconstructed using a weighted
combination of the eigenfaces. The projection weights form a feature vector for face
representation and recognition. When a new test image is given, the weights are computed by
Page | 23
projecting the image onto the eigen- face vectors. The classification is then carried out by
comparing the distances between the weight vectors of the test image and the images from the
database.
Back-Propagation Neural Network
The Back-Propagation is the best known and widely used learning algorithm in training
multilayer perceptrons (MLP) .The MLP refer to the network consisting of a set of sensory units
(source nodes) that constitute the input layer, one or more hidden layers of computation nodes,
and an output layer of computation nodes. The input signal propagates through the network in a
forward direction, from left to right and on a layer-by-layer basis. Back propagation is a multi-
layer feed forward, supervised learning network based on gradient descent learning rule. This
BPNN provides a computationally efficient method for changing the weights in feed forward
network.
5) Advantages:
BPNN with PCA algorithm is best as compare to PCA alone.
The BPNN algorithm is preferred over other neural network algorithms because of its
unique ability to minimize errors
6) Limitations: When number of subjects increases in recognition, execution time increases and
efficiency decreases
Page | 24
Literature Review 5
1) Paper Author : Navpreet Kaur
4) Work Description :
A system consisting of two phases which are the PCA pre-processing phase, and the
neural network classification phase. PCA is applied to calculate the feature projection
vector of a given face which is then used for face identification by the feed forward
neural network.
The proposed PCA and neural network based identification system provides improvement
on the recognition rates, when compared with a face classifier based on the PCA and
Euclidean Distance, all the data base will be created with a number of faces with different
makeup’s.
ICA- Independent component analysis (ICA) is a method for finding underlying factors
or components from multivariate (multidimensional) statistical data. There is need to
implement face recognition system using ICA for facial images having face orientations
and different illumination conditions, which will give better results as compared with
existing systems
5) Advantages :
LDA: The linear discriminant analysis (LDA) is a powerful method for face recognition.
It yields an effective representation that linearly transforms the original data space into a
low-dimensional feature space where the data is well separated
Page | 25
SVM: The advantage of SVM classifier over traditional neural network is that SVMs can
achieve better generalization performance.
ICA: What distinguishes ICA from other methods is that, it looks for component that are
both statistically independent and non Gaussian.
Literature Review 6
1) Paper Author : Kamini Solanki, Prashant Pittalia
Knowledge-based method
Knowledge-based methods are encoding our knowledge of human faces. These are rule-
based methods. They try to capture our knowledge of faces, and translate them into a set
of rules. It‟s easy to guess some simple rules. For example, a face usually has two
symmetric eyes, and the eye area is darker than the cheeks. Facial features could be the
distance between eyes or the colour intensity difference between the eye area and the
lower zone.
Feature-invariant methods
Facial recognition utilizes distinctive features of the face –including: distinct micro
elements like: Mouth, Nose, Eye, Cheekbones, Chin, Lips, Forehead, Ears, Upper
outlines of the eye sockets, the areas surrounding the cheekbones, the sides of the mouth,
and the location of the nose and eyes. The distance between the eyes, the length of the
nose and the angle of the jaw.
Page | 26
These algorithms compare input images with stored patterns of faces or features.
Template matching methods try to define a face as a function. One can try to find a
standard template of all the faces. Different features can be defined independently.
Appearance-based methods
A template matching method whose pattern database is learnt from a set of training
images. In general, appearance-based methods rely on techniques from statistical analysis
and machine learning to find the relevant characteristics of face images.
Advantages : .
LDA-based methods outperform PCA for both face identification and verification.
Limitations:
Illumination problem
Illumination problem happens when same image with condition. So person have to
keep with fix lighting condition, fixed distance, same facial expression and also same
view point. It can emerge extensively different when lighting condition is different.
Pose problem
Face recognition with different facial poses that is called pose problem. If face
rotation made very large changes in face appearance it reduce recognition rate. If
person try to match same image with different facial pose, it show the different result
Page | 27
3.2 Comparative Study:
Page | 28
efficinet decreases.
than only
PCA.
3. An 2014 2DPCA Image 2DPCA, Good Cannot deal
Integrated 2DLDA Covariance 2DLDA, efficiency directly with
Face Fuzzy set Image Fuzzy set, image data.
Recogniton Discrete Feature
Based on Matrix extraction Time
Wavelet consuming
Subspace. computation
for high
dimensional
data.
Page | 29
Page | 30
Chapter 4
Problem Statement:
For the past few years many approaches have been considered as far as face recognition is
considered. But almost all the approaches lack in some or the other aspects. According to
different research papers that have been reviewed the analysis shows that there are certain
limitations to the proposed algorithms.
Some of the encountered limitations include facial makeup which makes the face recognition
difficult. Any type of cosmetics used change the appearance of a person which in turn proves to
be a problem for the system to recognize the person.
Some systems are designed such that they only accept the frontal view of any image, so if there
is a side profile of the face available (left or right view of the face) then the system is not able to
identify the image for matching purpose
The next thing we see is that the system designed accepts only still images rather than supporting
real time environments so this also makes it an image dependent system, where the system will
have to rely on the type of image rather than a variety of images.
The other common issues include illumination problems which arise due to improper lighting
which may distort or blur the image, the pose problems which are caused by the improper setting
of the object before capturing the image. Also the background noise affects the quality of image
resulting in distortions and improper dimensions so that mathematical computations become
difficult
Chapter 5 Conclusion:
In this paper, we implemented the face recognition system using Principal Component Analysis
and Eigen face approach. The system successfully recognized the human faces and worked well
with different facial expressions and addition of salt and pepper noise. The face recognition
system was used to create an automatic attendance updating system based on face recognition
Page | 31
We have presented the definition, explanation and more details about face detection and
recognition in any image, so face images depend on the color mainly that uses color details,
connected component analysis. Our method uses compensates for the Lighting condition of the
image, and overcomes the difficulty of separating faces that are connected together using image
Feature based processing.Present study exposes that face recognition algorithm can be enhanced
using hybrid methods for better performance.
Face detection: For the constrained conditions, many face detection methods for static image are
not directly suitable to the task in video. We classified current approaches into groups, and
summarized their pros and cons. Face tracking: In face tracking head rotation and pose variations
are measure issues. Face tracking is a significant procedure in face recognition. It usually
exploits statistical model, exemplar-based model, and skin colour information to accomplish the
tracking task.
5.1 REFERENCES
[1] Delac K., Grgic M., Grgic S., \Independent Comparative Study of PCA, ICA, and LDA
on the FERET Data Set,International Journal of Imaging Systems and Technology,Vol.
15, Issue 5,2006, pp. 252-260
[2] Kara, S. and Hiremath, S. and Joshi, D.G. and Chadda, V.K. and Bajpai, A. \A Multi-
Algorithmic FaceRecognition System",International Conference on Advanced
Computing and Communications, 2006. ADCOM2006. pp 321 -326, 2006
[3] Text Book “Practical Image and Video Processing using MatLab” by OGE
MARQUES FloridaAtlanticUniversity[4] “A Survey of Face Recognition Techniques”
Rabia Jafri* and Hamid R. Arabnia* Journalof InformationProcessing Systems,
Vol.5,No.2, June 2009
Page | 32
[5] Face Recognition Methods & Applications Divyarajsinh N Parmar et al,
Int.J.ComputerTechnology &Applications, Vol4 (1),84-86
[6] R. Jafri, H. R. Arabnia, “A Survey of Face Recognition Techniques”, Journal of
Information ProcessingSystems, Vol.5, No.2, June 2009.
[7] C. A. Hansen, “Face Recognition”, Institute for Computer Science University of Tromso,
Norway.
[8] W. Zhao, R. Chellappa, P. J. Phillips & A. Rosenfeld, “Face recognitions literaturesurvey”,
ACM Computing
Surveys, Vol. 35, No. 4, December 2003, pp. 399–458.
[9] S. Suhas, A. Kurhe, Dr.P. Khanale, “Face Recognition Using PrincipalComponent Analysis
and LinearDiscriminant Analysis on Holistic Approachin Facial Images Database”, IOSR Journal
of Engineering e-ISSN:2250-
3021, p-ISSN: 2278-8719, Vol. 2, Issue 12 (Dec. 2012), ||V4|| PP 15-23
[10] M. A. Turk and A. P. Pentland, "Face Recognition Using Eigenfaces", 1991.[11]
Sukhvinder Singh, Meenakshi Sharma and Dr. N Suresh Rao, “Accurate FaceRecognition Using
PCA andLDA”, International Conference on Emerging Trends in Computer and Image
Processing (ICETCIP’2011)Bangkok December,2011
Page | 33