Human face detection is a significant problem of image processing and is usually a first step for face recognition and visual surveillance. This paper presents the details of face detection approach that is implemented to achieve accurate face detection in group color images which are based on facial feature and Support Vector Machine. In the first step, the proposed approach quickly separates skin color regions from the background and from non-skin color regions using YCbCr color space transformation. After the detection of skin regions, the images are processed with, wavelet transforms (WT) and discrete cosine transforms (DCT) as a result of which the 30×30 pixel sub images are found. These sub images are then assigned to SVM classifier as an input. The SVM is used to classify non-face regions from the remaining regions more accurately, that are obtained from previous steps and having big difference between faces regions and non-faces regions. The experimental results on different types of group color images show that this approach improves the detection speed and minimizes the false detection rate in less time and detects faces in different color images.

1. ACEEE Int. J. on Signal & Image Processing, Vol. 01, No. 03, Dec 2010
A Robust & Fast Face Detection System
Ritu Verma, Anupam Agrawal and Shanu Sharma
Indian Institute of Information Technology Allahabad, INDIA
Email: rituverma1021@gmail.com
Indian Institute of Information Technology Allahabad, INDIA
Email: {anupam@iiita.ac.in, shanu.sharma1611@gmail.com}
Abstract- Human face detection is a significant problem of color gives more reliability because it is not affected by
image processing and is usually a first step for face body posture and facial expression. It is easily
recognition and visual surveillance. This paper presents the distinguished from the background color. Hence the face
details of face detection approach that is implemented to detection approaches, based on the skin color, are widely
achieve accurate face detection in group color images which
used. But it is not sufficient to absolutely and precisely
are based on facial feature and Support Vector Machine. In
the first step, the proposed approach quickly separates skin detect the face only by using skin color information. When
color regions from the background and from non-skin color several faces are very near to each other or the face regions
regions using YCbCr color space transformation. After the and other body regions are close or skin-likelihood
detection of skin regions, the images are processed with, background is connected together to the face, it often
wavelet transforms (WT) and discrete cosine transforms increases the false detection ratio. This problem can be
(DCT) as a result of which the 30×30 pixel sub images are handled by detecting the false candidate regions with
found. These sub images are then assigned to SVM classifier statistical methods. In this face detection system the sub
as an input. The SVM is used to classify non-face regions from images of faces are very small in size for which the
the remaining regions more accurately, that are obtained
statistical learning is used. Statistical learning theory is
from previous steps and having big difference between faces
regions and non-faces regions. The experimental results on currently the best theory for small samples statistics
different types of group color images show that this approach estimates and projection learning. SVM theory is
improves the detection speed and minimizes the false established on the basis of statistical learning theory; its
detection rate in less time and detects faces in different color objective is to resolve the problem of classification of small
images. samples.
Index Terms: Face Detection; Skin Color Detection; Wavelet The outline of the paper is prepared as follows: The
Transform; Discrete Cosine Transform; Support Vector summary of literature survey described which is similar to
Machine. my system and few face detection methods with their
merits and demerits. Section III explains the details of the
I. INTRODUCTION implementation and methods we have been used. In section
IV the results of this face detection approach on various
A face detection system is a system that determines the types of images are discussed and in section V the
locations and sizes of human faces in arbitrary (digital) conclusion and scope for the future work are explained.
images. It detects facial features from images and ignores
all other things, like buildings, trees etc. Recently, II. RELATED WORK
researchers have proposed to detect face by method
combining features and color to obtain a high performance Face detection technique is an open challenge from last
and high speed results [1], [4] and [13]. Detecting faces is a many years, and various solutions addressing the face
crucial step in the identification applications for example detection problem have been proposed under different
airport security, law enforcement etc. Most of the face categories which are discussed below. Face detection is not
recognition and face tracking algorithms assumes that the an easy method as the detection is affected by many
initial face localization is known. The main merit of any internal and external factors.
good approach is to provide fast and high detection ratio Few main Face Detection Methods are as follows:
and can deal with faces in complex background. A. Knowledge-Based Method:
In this paper, implementation of a robust face detection
algorithm which is based on facial feature and LSVM In this method the relationship between facial features
(linear support vector machine) is presented. This of test image is used to represent the content of the face and
algorithm deals with different complexities and provides then encode picture digitally as a set of rules and to reach
high speed and high detection ratio. Different complexities the finest scale. It is a top down approach [5]. Merits and
include finding number of faces in group image, varying demerits of knowledge-based method are as follows:
illumination, occlusion and complex background present in Merits
an input image.
• It is simple to describe the features of face and their
The skin color is a significant feature of a face. It has a
relationship by using simple rules.
strong cluster feature of YCbCr and HIS color space [1]. In
• By coded rules first facial features of image are extracted
YCbCr, Y stands for the “luma” (luminance) which is
then candidate faces are identified.
brightness. Cb and Cr stand for the “color difference” of
blue – luma (B-Y), and red – luma (R-Y) respectively. Skin
© 2010 ACEEE 17
DOI: 01.IJSIP.01.03.135

2. ACEEE Int. J. on Signal & Image Processing, Vol. 01, No. 03, Dec 2010
Demerits Demerits
• Translation of human knowledge into precise rule is very • Difficult to locate facial feature due to various
difficult. complexities (illumination, occlusion etc.) in an image.
• General rules may find many false positives. • Difficult to detect features in complex background.
B. Template Matching Method: D. Appearance-Based Method:
This method is based on finding the co-relation between This method learns the templates from the set of
a test sub image and the pre-defined stored face patterns. training images. It finds the relevant characteristics of face
The predefined images might be the whole face or and non-face by using statistical analysis and machine
individual face features such as nose, eyes, mouth, learning techniques [3] and [7].
eyebrows, and lips [5]. Algorithms used under this method are:
Algorithms used under this method are:
Eigen Faces:
Predefined Face Templates: These are also called the eigenvectors, in which
In predefined face templates several templates for the different algorithms are used to approximate the
whole/individual or both parts (whole & individual) of the eigenvectors of the auto correlation matrix of a candidate
face are stored. image [19].
Deformable Templates: Neural Network:
In this an elastic facial feature model as a reference A network of neurons (simple element) called nodes
model is stored and the deformable template mode of the used is to perform function in parallel. Central nervous
object of interest is fitted in. system gave this idea of neural network. These networks
Merits and Demerits of Template Matching Method are as are trained for the detection of faces by providing it, face
follows: and non-face samples [15].
Merits Support Vector Machine:
• It’s simple and easy to implement. Support vector machine are learning machine and it
makes binary classification. The idea is to enlarge the
Demerits difference or margin between the vectors of negative and
• Templates have to be initialized near the face images. positive sets and obtain an optimal boundary which
• Difficult to enumerate templates for different poses. separates two sets of vectors [8] and [14].
C. Feature-Invariant Approach: Hidden Markov Model:
In this approach faces structural features are not It is also abbreviated as HMM model and can be
changed under different conditions, such as varying considered as simple dynamic Bayesian network. Hidden
viewpoints of cameras, pose angles, and /or illumination Markov Model is a class of statistical model which uses the
conditions. statistical properties of a signal that model the processed
Algorithms used under this approach are: system. The Markov parameters should be taken from the
observed parameters [16].
Colour-Based Approach:
Merits and demerits of Appearance-based method are as
Colour based is also called skin-model based method. This follows:
approach is based on the fact that different skins from
different races are clustered in a single region and makes Merits
use of the skin colour as indication to the presence of • Use powerful machine learning algorithms and it has
human beings [1], [4] and [6]. demonstrated good empirical results.
Facial-Feature Based Approach: • It offers to detect faces in various poses and orientations.
In this method global and/or detailed features are used Demerits
for face detection. It has become popular in present days. • It is usually needed to look for the space and scale.
The global features (e.g. skin, size and shape) are firstly • It requires lots of positive and negative examples.
used to detect the candidate area after that they are tested
using detailed features (e.g. eyes, nose, and lips) [13]. II. DETAILS OF THE APPROACH IMPLEMENTED
Merits and Demerits of Feature-invariant approach are as
follows: The flow chart of a proposed approach is shown in
figure1.
Merits
• Features are invariant in different poses and orientations
of the faces.
© 2010 ACEEE 18
DOI: 01.IJSIP.01.03.135

3. ACEEE Int. J. on Signal & Image Processing, Vol. 01, No. 03, Dec 2010
color space. Segmented skin color regions are obtained by
Input the elliptical cluster method for the skin tones in the
imag Color Morpholo Discrete
Space gy Based Wavelet transformed YC’bC’r space. It is described in equations (1)
e
Based Operation Transfor and (2) as given below [7].
Segmentat m
Outp … (1)
ut Classificati Discrete
imag on by Cosine … (2)
SVM Transfor
m
Where a = 25.39, b = 14.03, ecx = 1.60, ecy = 2.41, Ѳ =
2.53, cx =109.38, and cy =152.02 are computed in the
YC’bC’r space [7].
Figure 1: Flow chart of the approach used for face detection
The images are received after lighting compensation
technique, and are filtered with a 3×3 low pass filter [18]
Steps for Face Detection: which is used for minimizing the effect of noise. If then the
1. First give a RGB image as an input image to the Skin pixel satisfies equation (1) in elliptical cluster method
color model. (YC’bC’r color space), it is marked as 1 and has to be
2. The Skin color model converts the RGB image to the considered as skin color pixel. Otherwise, it is marked 0
YCbCr color space model [18]. and has to be considered as non-skin color pixel. It
3. For handling varying lighting conditions convert this provides an output binary image after the above process.
output image in YC’bC’r color space by the elliptical Finally it can detect skin color regions accurately after
formula [7]. morphological (dilation) operation [18].
4. For reducing noise effects filter this image by 3×3 low
a. B. Discrete Wavelet Transform:
pass filter, and then apply morphology (dilation)
operation to get a binary image [18]. For reducing the training time and SVM dimension, the
5. Find the skin regions based on above binary image. samples are compressed by wavelet transform (WT). Here
6. The discrete wavelet transform (DWT) decomposes the using the discrete wavelet transforms which is based on
given input image into a set of sub-bands of different sub-band coding and it is found to create a fast computation
resolutions and selects the low frequency parts. The of WT [12]. It is easy to execute and minimize the
new generated top left low frequency sub-bands are computational time and resources required.
nearly equal to the original image [18]. The discrete wavelet transform decomposes the input
7. Take the output of the DWT to the DCT and use 30×30 frame of image into a set of sub band of different
size window to pick up the significant information of resolutions. The new generated sub-band is nearly equal to
signal energy [11]. the original frame. DWT is a time-scaled representation of
8. Support Vector Machine is used for classification to the digital signal and is found by digital filtering techniques
construct an optimal hyper-plane which has a maximum [18]. The amount of the information present in the signal is
margin of the separation between the face and non-face measured and this is termed as the resolution of the signal
classes [8]. We have taken 30×30 size of windows as an which is to be finding out by several filtering operations
input and separate these in faces or non-faces by the and it is given by up-sampling and down-sampling
classification. phenomena. The dilation function of discrete wavelet
9. Obtain the final face detected output image. transform is represented by a tree of low & high pass
Details of main components of the approach are given filters. Low pass filters are transforming in each step. The
below: original signals are continuously decomposed into the
subpart of lower resolution and the high frequency
A. Skin Color Model And Segmentation: components are not analyzed.
In order to apply this method in the real time system, Wavelet coefficients are created into wavelet blocks in
skin color detection is adopted; de-noising and lighting which horizontal, vertical and diagonal edges are the sub
compensation are the initial steps of skin color model. This images of real image, it is shown in figure2. The upper
is because the lighting condition and noise has great effect most left sub image represents the superior level of low
on the skin color detection. YCbCr color space pass sub image. The concept of wavelet block gives an
transformation is faster than the other approaches and association between coefficient and what they represent
popularly used in skin color detection [2]. YCbCr color spatially in the frames [10].
space is developed for television systems, and it is
luminance separated color space so it is widely used in
mpeg, jpeg and other video compression standards.
First linear conversion of RGB color space to YCbCr
color space is obtained, but for further reduction in the
lighting effect and to obtain a good result of skin color
cluster, a segmented non-linear conversion algorithm [7] is Figure2: wavelet block are reconstruction of wavelet coefficient.
used which converts YCbCr color space into the YC’bC’r This is a four level discrete wavelet transform [10]
© 2010 ACEEE 19
DOI: 01.IJSIP.01.03.135

4. ACEEE Int. J. on Signal & Image Processing, Vol. 01, No. 03, Dec 2010
C. Collecting Training Sample: linear problem. A LSVM classifier is designed to classify
In the previous methods training samples are collect and used LibSVM [8] to train the samples. The LSVM
from the database directly and the non-face samples are kernel function is adopted here-
selected from the scenery images, such as building, plants, K (xi , xj) = < xi , xj > …..(3)
trees and so on. So that it narrows the selecting scope. But
here the training samples are selected after the processing In a binary classification with l sample points:
with color transform, de-noising, and detection of skin
color regions and so on. Here we use 12 images for testing (xi , yj) i = 1,2,3…………..l …..(4)
purpose which are collecting from personal digital camera
and also from the database [17]. After the initial steps like- Where xi є Rn and yi = {+1, -1} are the classifying label
color space transformation, lighting compensation and [7]. This system finds faces by thoroughly scanning an
detection of skin regions we get scaled images. From the image for face like patterns at several possible scales, by
scaled images we extract 30×30 pixel sub-images and here isolating the original image into overlap sub-images and
we get around 700 sub-images from 12 testing images and determines them into appropriate class face or non-face by
extract them in 150 faces and 550 non-faces. using support vector machine. The figure 3 shows the
geometrical interpretation of the technique support vector
D. Discrete Cosine Transform:
machine provides in the framework of the face detection.
The DCT is a good example of the transform coding The vital use of support vector machine is in the
[18]. The recent JPEG standard images use the DCT as its classification step, which is the essential part of the work.
basis. The discrete cosine transform relocates the high By using support vector machine classify all window
valued energies (information) to the upper left corner to the patterns and if the class matches a face then make a square
image and the lesser energies are relocated in other areas around the face in the output image.
[11]. Discrete cosine transform is a unique method that has
near-optimal energy compaction property [9]. It separates
the given image into sub–bands (parts of image) on the
basis of visual quality. The DCT has a great feature
extraction and excellent data compression and has less Non-
Faces
computing features. It gives robustness for detection in
lighting effects or variations.
Figure3: SVM separate the face and non-face by geometrical
Energy Compaction is the main property of DCT [11]. interpretation. The patterns are real support vectors obtained after
Having a power to produce a transformation scheme can be training the system [8]
directly approximated by its ability to compact input data
Faces
into a few possible coefficients. It allows quantizer to
remove coefficient with relatively small amplitudes and IV. EXPERIMENTAL RESULTS
reconstruct image without any visual distortion. DCT
exhibits excellent energy compaction for highly correlation Here evaluation of proposed methodology on a face image
sub-images. In the transform coding, the pixels in an image database, and construction of the database for face
displays a certain level of correlation with neighboring detection from personal photo collections and internet [17]
pixels. Same problem is there in video transmission which is done. These color images or the database has been taken
shows very high correlation of adjacent pixels in under different complexities, like detecting possible faces
consecutive frames. We take the output of Discrete under varying illumination conditions and occlusion in
Wavelet Transform as an input to the Discrete Cosine group photographs with complex backgrounds. With high
Transform and use 30×30 size window to pick out the detection rate of 87.65% accuracy, this approach can detect
significant information of signal energy. The sample all possible faces in between range (9.38sec to 11.97sec) of
feature vector is extracted and compacted by DCT [7]. time. The face detection time depends on the complexities
of the testing color images. Further the discussed approach
E. Support Vector Machine: is able to detect multiple numbers of faces with broad range
A SVM is a supervised learning technique form of of facial variations in an image.
machine learning, and it is applicable for classification and
regression. This support vector machine theory is A. Discussion for the output images shown in section B are
developed by Vladimir Vapnik & his team in 1995 at AT& given below:
Bell Laboratories, and the principle is based on structural 1. The first input image is the original RGB image which
risk minimization, so it has very good generalization ability we get either from the personal dataset or from the
[8]. Generalization means the summation of data and internet datasets [17], having different complexities.
knowledge. For example the given input image1 has varying
The main aim of statistical learning theory is to present illumination over different faces and has complex
a framework for studying the problem of inference, which background.
is of gaining knowledge, making predictions, making 2. Perform low pass filtering to reduce effect of noise and
decisions or constructing models from a set of data. The for handling varying lighting condition use elliptical
proposed method adopts a kernel function so it is able to formula (as discussed in above) on the input image.
solve the dimension problem, and is well suited for non- From this we get the binary skin map image.
© 2010 ACEEE 20
DOI: 01.IJSIP.01.03.135

5. ACEEE Int. J. on Signal & Image Processing, Vol. 01, No. 03, Dec 2010
3. Third image shows the skin region detected image of Complexities in different input images which are shown
the input RGB image. Here we separate the background in below section B and section C are:
of image from the skin color regions. 1. Image1 has complexity of varying illumination over
4. For the fourth image, perform the dilation operation different faces and has complex background (skin
(morphological operation) on the 2nd skin map region likelihood background).
image. The dilation operation which accepts the 2. Image9 has complexity of occlusion and has complex
structuring element objects, known as STRELs [18]. background.
5. The fifth image shows the dilated skin region detected 3. Image10 has complexity of tilted faces.
image of the input image after applying the above
B. The output images (2 to 8) generated by various steps
operations on the 4th image.
on input image (1) are given below:
6. Apply discrete wavelet transform to get a sixth scaled
image.
7. After getting the scaled image apply discrete cosine
transform. By applying this process the image is
divided into the 30×30 sub-images, and we train all
sub-images as a face or non-face sub-image.
8. In seventh image, Support Vector Machine (SVM) is
used for classification of data to construct an optimal
hyper-plane which has a maximum margin of the
separation between the face and non-face classes.
9. Finally we obtain the final face detected output image Image1. The original RGB Image2. Skin map image
(image8) after classification, where faces are enclosed image
in boxes around them.
Here, we have collected 12 testing color images of
different sizes and different complexities. In these 12
testing group color images, first six images (1 to 6) are
taken from personal digital camera and the next six images
(7 to 12) are taken from the face detection datasets “Bao
Face Database” [17]. Total 81 faces are there in 12 images
in which 71 faces are detected successfully. This approach
gives accuracy 87.65% with a good speed. After the Image3. Skin region detected Image4. Dilated skin map
training time of the faces and non-faces it can able to detect image image
the possible faces in between range 9.38sec to 11.97sec. Its
detection timing depends on the complexities of the
images. Table1 and Table2 show the results of finding
faces in different given input images.
TABLE I:
FACE DETECTION RESULTS ON THE PERSONAL SIX (1 TO 6)
TESTING COLOR IMAGES.
Sr. Number of Correct Missing Detection
no. faces in detection of detection time of Image5. Dilated skin region Image6. Scaled image after
images faces of faces faces(sec) detected image applying DWT image
1 6 6 0 9.87
2 6 6 0 10.16
3 6 6 0 9.77
4 5 5 0 9.64
5 6 6 0 9.38
6 4 3 1 11.97
TABLE II:
FACE DETECTION RESULTS ON THE DATEBASE SIX (7 TO 12)
TESTING COLOR IMAGES.
Sr. Number of Correct Missing Detection
Image7. Classification by Image8. Final face detected
no. faces in detection of detection time of
SVM image image
images faces of faces faces(sec)
7 12 8 4 10.24
8 9 6 3 9.99
9 8 8 0 9.96
10 5 5 0 10.88
11 7 5 2 11.44
12 7 7 0 11.20
© 2010 ACEEE 21
DOI: 01.IJSIP.01.03.135

6. ACEEE Int. J. on Signal & Image Processing, Vol. 01, No. 03, Dec 2010
C. The output for more images with different complexities: [4] Yepeng Guan and Lin Yang, "An unsupervised face
detection based on skin color and geometric information," Sixth
International Conference on Intelligent Systems Design and
Applications, (ISDA'06), Volume 2, 2006, pp. 272-276.
[5] Muhammad Usman Ghani Khan and Atif Saeed,
“Human detection in vedios,” Journal of Theoretical and Applied
Information Technology, 2009, pp. 212-220.
[6] C. Garcia, G.Zikos and G.Tziritas, “Face Detection in
Color Images Using Wavelet Packet Analysis,” IEEE Conference
on Multimedia Computing and Systems, volume 1, 1999, pp. 703-
708.
[7] Jinxin Ruan and Junxun Yin, "Face Detection Based on
Image9. Face detected Image10. Face detected
image image
Facial Features and Linear Support Vector Machines,"
International Conference on Communication Software and
Networks, 2009, pp. 371-375.
V. CONCLUSION AND FUTURE WORK [8] C. C. Chang and C.J. Lin. “LIBSVM - A Library for
support Vector Machines” Available
This paper discusses a robust & fast face detection on:http://www.csie.ntu.edu.tw/~cjlin/libsvm/ 2008. (Accessed
approach and its implementation is based on facial feature 10th Feb 2010).
and LibSVM. The statistical learning theory is related to [9] Chai Beng Seow, Regina Gani and Varun Jeoti,
the training samples. Selected samples and regions which “Wavelet-DCT based image coder for video coding applications,”
are found from the skin color regions by non-linear International Conference on Intelligent and Advanced Systems,
conversion are used; the strength of samples and the ICIAS, 2007, pp.748-752.
functioning or the performance of classifier is improved. [10] J. Karlekar and U. B. Desai, “Finding faces in color
images using wavelet transform,” International Conference on
For the compression purpose we use here discrete wavelet
Image Analysis and Processing, 1999, pp. 1085-1088.
transform and for extracting the feature vector of sample [11] “The Discrete Cosine Transform (DCT): Theory and
images we use discrete cosine transform, so the resultant Applications”, Available on:
matching time and the training difficulty of support vector http://www.wisnet.seecs.nust.edu.pk/publications/tech_reports/D
machine are obviously reduced and there is speeding up the CT_TR802.pdf (Accessed 15th Mar 2010).
algorithm. Result shows that the algorithm achieves good [12] R. de Queiroz, C. K. Choi, Y. Huh and K. R. Rao,
(around 87.65%) detection accuracy, lower false detection “Wavelet Transforms in a JPEG-Like Image Coder,” IEEE
rate and improved speed, which makes the algorithm Transaction on Circuit and Systems for Video Technology,
highly robust. volume 7, no. 2, April 1997, pp. 419-424.
[13] Tse-Wei Chen, Shou-Chieh Hsu and Shao-Yi Chien,
Further the present work may be extended to reduce the
"Automatic Feature-Based Face Scoring in Surveillance
false detection rate, solve the problem of shifted boxes and Systems," Ninth IEEE International Symposium on Multimedia,
improve its accuracy for face recognition. ISM, 2007, pp. 139-146.
[14] E. Osuna, R. Freund and F. Girosi, “Training Support
ACKNOWLEDGEMENT Vector Machines: An Application to Face Detection,” IEEE
Computer Society Conference on Computer Vision and Pattern
The authors would like to express sincere gratitude to Recognition, 1997, pp.130-136.
the Director of the Institution Dr. M.D Tiwari for providing [15] L. Mostafa and S. Abdelazeem, “Face Detection based
excellent computational facilities and stimulating work on Skin Color using Neural Networks,” First ICGST International
environment for carrying out the research work. Conference on Graphics, Vision and Image Processing GVIP
2005, pp. 53-58.
REFERENCES [16] Yi-Qiong, Bi-Cheng Li and Bo Wang, “Face detection
and recognition using neural network and hidden markov
[1] R.L. Hsu, M. Abdel-Mottaleb and A.K. Jain, "Face Detection models,” International conference on neural networks and signal
in Color Images," IEEE Transactions on Pattern Analysis and processing, 2003, pp. 228-231.
Machine Intelligence, volume 24, no. 5, May 2002, pp. 696-706. [17] Face detection datasets “Bao Face Database”, Available
[2] Yu-Ting Pai, Shanq-Jang Ruan, Mon-Chau Shie and on: http://www.facedetection.com/facedetection/datasets.htm
Yi-Chi Liu, “A Simple and Accurate Color Face Detection (Accessed 10th May 2010).
Algorithm in Complex Background,” IEEE International [18] Rafael C. Gonzalez, Richard E. Woods and Steven L
Conference on Multimedia and Expo, ICME-2006, pp. 1545- Eddins, “Digital Image Processing using MATLAB”, Pearson
1548. Education Inc. 2005.
[3] Michael J. Jone and Paul Viola, “Fast Multi-view Face [19] Hossein Sahoolizadeh and Youness Aliyari Ghassabeh,
Detection,” TR2003-96 August 2003. Available on: “Face recognition using eigen-faces, fisher-faces and neural
http://www.merl.com/reports/docs/TR2003-96.pdf. (Accessed on networks,” 7th IEEE internet conference on cybernetic intelligent
5th Feb 2010). systems, 2008, pp
© 2010 ACEEE 22
DOI: 01.IJSIP.01.03.135