Published by : International Journal of Engineering Research & Technology (IJERT)

http://www.ijert.org ISSN: 2278-0181

Vol. 8 Issue 05, May-2019

Face Recognition System

Shivam Singh Prof. S. Graceline Jasmine
Department of SCSE Department of SCSE
Vellore Institute of Technology, Vellore Institute of Technology,
Chennai Tamil Nadu, India Chennai Tamil Nadu, India

Abstract:- In present times, face recognition has become one of use some system algorithms to pick out specific,
the best technologies for computer vision. Face recognition is distinctive details about a human’s face.
always a very difficult task in computer vision, illumination,
pose, facial expression. Face recognition tracks target objects 2.2. Face Detection
in live video images taken with a video camera. In simple
words, it is a system application for automatically identifying
In [1] This face detection process actually verifies the
a person from a still image or video frame. In this paper we image is face image or not. Detection process actually
proposed an automated face recognition system. This works on Haar Cascade classifier. Object Detection using
application based on face detection, feature extraction and Haar feature- based classifiers is an effective object
recognition algorithms, which automatically detects the detection method proposed Paul Viola and Michael Jones.
human face when the person in front of the camera It is machine learning based approach where a cascade
recognizing him. We used KLT Algorithm, Viola-Jones function is trained from images. It is used to detect objects
Algorithm face detection which detect human face using Haar in other images.
cascade classifier, however camera is continuously detecting
the face every frame, PCA algorithm for feature selection. We
apply a model combining to match the geometric 2.3. Haar Cascade Classifier Features
characteristics of the human face. In [2] Here we calculated, the first feature selected seems
Keywords—Face Recognition, Face Detection, PCA.
to focus on the property that the region of the eyes in often
darker than the region of the nose and cheeks. The second
feature chosen is based on the eye is darker characteristics
1. INTRODUCTION
than the bridge of the nose. However, you do not need the
same window that applies to your cheeks and other places.
Human Face always play crucial role in application such as
security system, credit and debit card verification
surveillance on identify criminal public places. The main
objectives of the system are to create a facial recognition
system that can be emulated and eventually overcome this
capacity of human. This system focuses especially on the
human frontal faces. Multiple face recognition algorithms
Figure 1: Haar Cascade Classifier
have been developed and each has its own strength. Most of
the time we look at a face and are able to recognize it
instant if we are already familiar with the face. This natural face recognition system that does capturing the image of
ability, if possible, can be justified and can be used for real- face feature detection, extraction, storing and matching.
life applications. That time there are many face detection But the difficulty occurs to lay the transmission lines in the
algorithms. The first one is a local face recognition system, places where the topography is bad. The authors proposed
which uses facial features of a face to intimity the face with a system based on real-time face recognition that is
a person. The second approach or global face recognition reliable, secure and fast, and requires improvement in
system use the entire face to recognize a person. The above different lighting conditions.
two process have been implemented one to another way by 3. PROPOSED WORK
another algorithms. The neural network and its feasible Systems design is a process that defines architecture,
applications in the field of research.The complications of a components, modules, interfaces, and data requirements.
facial features that take place over time. Unconcern of those Figure [2] System design can be viewed as a system theory
changes can easily identify a person. So, the idea of application for product development. The face detection
emulate this skill is that human beings can be very technology that helps locate human face in digital images
rewarding. and video frames. The object detection technology that
2. RELATED WORK deals with detecting instances of objects in digital image
2.1. Face Tracking and videos. The proposed automated recognition system
The objective of this algorithm is to detect object of face in can be divided into five main modules:
real time and to keep tracking of the same object. Here we
3.1. Image Capture
use the training samples images of other objects of your
A camera is placed away from the entrance to capture an
choice to be detect and track by training classifier. Face
image of the front of the student. And a further process
tracking is a part of face recognition system. Here we can
goes for face detection.

IJERTV8IS050150 www.ijert.org 263

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijert.org ISSN: 2278-0181
Vol. 8 Issue 05, May-2019

3.2. Face Detection and Facial Features 3. Use viola Jones and KLT Algorithm Extract the
The appropriate and effective facial detection algorithm Region of interest in Rectangular Bounding Box.
constantly improves facial recognition. Several facial
algorithms such as face-to-face geometry, construction 4. Convert to gray scale, apply histogram equalization
methods, Face geometry-based methods, Feature Invariant and Resize to 100x100 i.e. Apply pre-processing
methods, 5. if Enrollment Phase
then Store in Database
else
Apply PCA (For feature Extraction)
end if

6. Post-processing

4. FEATURE SELECTION AND EXTRACTION

Throughout the past few decades there have been

many face detection techniques proposed and
implemented. Some of the common methods described
by the researchers of the respective fields are:

Figure 2: System Diagram

Machine learning based methods. Out of all these methods

Viola and Jones proposed a framework that gives a high
detection rate and is also fast. Viola-Jones detection
algorithm is fast and robust. So we chose Viola-Jones face
detection algorithm, which uses Integral Image and
AdaBoost learning algorithm as classier. We have
observed that this algorithm yields better results in a
variety of lighting conditions.

3.3. Pre-Processing
Figure 3: Extracting the face features
Extracting the face features it is called pre-processing. This
pre-processes step involves specifying the extracted facial
image and transforms to 100x100. Histogram Equalization formulate relevant tags and categories. We can then
is the most commonly used Histogram Normalization calculate the feature vectors for each of the training
technique. This improves the contrast of the image as it images, and test image, take their dot products and return
extends beyond the intensity of the image, making it even the one with the highest dot product as the match.
more clear and constraint.
4.1 Principal component analysis (PCA)
3.4. Database Development In [3] Face Detector, for which several non-derived
As we choose biometric based system every individual is learning methods are available. These include, for
required. This database development phase consists of an example, the OpenCV based face detectors, and the Haar
image capture of each individual and extracting the bio- Cascades.Elaborate work by Viola and Jones, while later
metric feature, and then it is enhanced using pre- based on Gradiente's histogram. PCA is used to describe
processing techniques and stored in the database. face images in terms set of base functions, or eigenfaces.
Eigenface was described in early identification problems.
3.5. Post-Processing PCA is a technique, so the process does not rely on class
In the proposed system, after recognizing the faces of the definition. In our implementation of eigenvalues,
person, the names are show into a video output. The result Euclidean distance. Multiple linear principal components
is generated by exporting mechanism present in the analysis. However, a face picture and video are a
database system. These generated records can be seen in multilinear array, this vector define a 1D vector from the
real time video. This ensures that person whose faces are face image and liner projection for the vector. I think it can
not recognized correctly by the system have to check in help for optimization to classify the face pixels. Ever
database. And thus, giving them the ability to correct the consider Use of eigenfaces for dimensional reduction of
system and make it more stable and accurate. eigen values, and fisher faces for feature extraction (linear
discriminant analysis). Face Detector, for which several
3.6. Proposed Algorithm simpler (non-deep learning) techniques are available.
1. Capture the Person’s Image. These include, for example, the OpenCV based face
2. Apply Face detection algorithms to detect face. detectors, and the Haar Cascades, a seminal work by Viola

IJERTV8IS050150 www.ijert.org 264

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijert.org ISSN: 2278-0181
Vol. 8 Issue 05, May-2019

and Jones, while the latter is based on the Histogram of The system has created a database by own then you do
Gradients. stuff with your rectangles that highlight the faces, as such:

Figure 4: Check the face in the data base

Figure 6: Training Database

4.2 Neural Networks

In [4] Machine learning approaches to image recognition Therefore, it could be said that the image of the original
involve identifying and extracting important features from face can be reconstructed from the own interfaces if the
images and using them as input to a machine learning proper characteristics (faces) are added in the correct
model. Image Rrecognition is a machine learning method, proportion. Each face represents only certain features of
designed to mimic the way the human brain works. the face, which may not be present in the original image.

5 RESULTS
With this method, the computers are taught to recognize
the visual elements within by relying on large databases and
noticing emerging patterns, the computers can make sense
of images. If the similarity is below a threshold, you can
return ‘not matched’ as well.

Figure 7: Database Size and Recognize Rate

 Demonstration of different vision systems
 Face detection using Haar-Cascades
database contains a distinct image. Face recognition using
The use of neural networks for face recognition has been eigenface and LBP pattern hisrogram[8].
shown by and we can see the suggestion of a semi-
 Face recognition using Eigenface face recogniser
supervised learning method that uses support vector
 Face recognition using Local binary pattern
machines for face recognition. The Recognition system is
histograms
simple and works efficiently.

The performance of this method is compared with other

existing face recognition methods and it is observed that
better accuracy in recognition is achieved with the
proposed method. Face Recognition using KLT algorithm
[5] and Fusion of PCA and recognition plays a vital role in
a wide range of applications. It is high rate accuracy
applications in identifying a person is desired. This

IJERTV8IS050150 www.ijert.org 265

(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijert.org ISSN: 2278-0181
Vol. 8 Issue 05, May-2019

[3] Ohol, M. R. M., & Ohol, M. S. R. PCA Algorithm for Human Face
Recognition.
[4] Kasar, M. M., Bhattacharyya, D., & Kim, T. H. (2016). Face
recognition using neural network: a review. International
Journal of Security and Its Applications, 10(3), 81-100.
[5] Mikhaylov, D., Samoylov, A., Minin, P., & Egorov, A. (2014,
November). Face Detection and Tracking from Image and
Statistics Gathering. In Signal-Image Technology and Internet-
Based Systems (SITIS), 2014 Tenth International Conference on
(pp. 37-42).
[6] Liu, Z., & Wang, Y. (2000). Face detection and tracking in
video using dynamic programming. In Image Processing,
2000. Proceedings. 2000 International Conference on (Vol. 1, pp.
53-56). IEEE.
[7] IEEE ,Boda, R., & Priyadarsini, M. J. P. (2016). Face Detection
Figure 8: Calculate the number of Components and Tracking Using KLT And Viola Jones. ARPN Journal of
Engineering and Applied Sciences, 11(23), 13472-1347
[8] M. Turk and A. Pentland, Eigenfaces for recognition, Journal of
6 CONCLUSIONS Cognitive Neuroscience, 3(1), pp. 7186, 1991.
[9] H. Lu, K. N. Plataniotis, and A. N. Venetsanopoulos, Mpca:
Multilinear principal component analysis of tensor objects, IEEE
In this paper, after experimenting several techniques all Trans. on Neural Networks, 19(1):1839,2008.
technique is working well face recognition. Face [10] Harguess, J., Aggarwal, J.K., A case for the average- half-face in
Recognition Systems is based on face recognition. This 2D and 3D for facerecognition, IEEE Computer Society.
system can be used to identify unknown person. In real-
time scenarios, PCA outperforms other algorithms. The
future work is for the recognition of the algorithm. In [10],
the system developed only by recognizing the 30-degree
angle variations that should be improved. Gait recognition
can be fused with face recognition systems. Poor lighting
conditions. Our system will perform well but it is not a
perfect solution.

7 ACKNOWLEDGEMENT

With all respect and gratitude, we would like to thank all

the people who have helped us directly or indirectly for the
completion of the project "Face Recognition System ". We
express our heartily gratitude towards Prof.S. Graceline
Jasmine for guiding us to understand the work
conceptually and also for her constant encouragement to
complete the project.

REFERENCES

[1] Deshpande, N. T., & Ravishankar, S. (2017). Face Detection and

Recognition using Viola-Jones algorithm and Fusion of PCA and
ANN. Advances in Computational Sciences and Technology,
10(5), 1173- 1189.
[2] Kavia, M. Manjeet Kaur, (2016). “A Survey paper for Face
Recognition Technologies”. International Journal of Scientific and
Research Publications, 6(7).

IJERTV8IS050150 www.ijert.org 266

(This work is licensed under a Creative Commons Attribution 4.0 International License.)