Analysing Assorted Window Sizes with LBG and KPE Codebook Generation Techniques for Grayscale Image Colorization

Shared by: ijcsiseditor

Stats

views:: 88
posted:: 7/5/2011
language:: English
pages:: 5

Public Domain

Document Sample

(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 6 June 2011

Analysing Assorted Window Sizes with LBG and
KPE Codebook Generation Techniques for Grayscale
Image Colorization
Dr. H. B.Kekre Dr. Tanuja K. Sarode Sudeep D. Thepade Ms. Supriya Kamoji
Sr. Professor,MPSTME, Asst. Professor, Asst. Professor, Sr.Lecturer,
NMIMS Deemed-to-be Thadomal Shahani Engg. MPSTME, Fr.Conceicao Rodrigues
University,Vileparle (W), College, NMIMS Deemed-to-be College of Engg,
Mumbai-56, India. Bandra (W), Mumbai-50, University,Vileparle (W), Bandra (W),
India. Mumbai-56, India. Mumbai-50, India.
Abstract—This paper presents use of assorted window sizes and Gray scale image is represented by only the luminance values
their impact on colorization of grayscale images using Vector that can be matched between the two images. Because a single
Quantization (VQ) Code Book generation techniques. The luminance value could represent entirely different parts of an
problem of coloring grayscale image has no exact solution. image, the remaining values within the pixel’s neighborhood
Attempt is made to minimize the human efforts needed in
manually coloring grayscale images. Here human interaction is
are used to guide the matching process. Once the pixel is
only to find reference image of similar type. The job of matched, the color information is transferred but original
transferring color from reference image to grayscale is done by luminance value is retained [2].
proposed techniques. Vector quantization algorithms Linde Buzo
and Gray Algorithm (LBG) and Kekre Proportionate Error The details in color image can be utilized for analysis and study
(KPE) are used to generate color palette in RGB and Kekre’s LUV of particular image in the applications like medical
color space. For colorization source color image is taken as
tomography, information security, image segmentation, etc.
reference image which is divided into non overlapping pixel
windows. Initial clusters are formed using VQ algorithms LBG Coloring of old Black and White movies and rare images of
and KPE, used to generate the color palette. Grayscale image monuments, celebrities is one of the best applications which
which is to be colored is also divided in non overlapping pixel give good feel and understanding.
windows. Every pixel window of gray image is compared with
color palette to get the nearest color values. Best match is found In case of pseudo-coloring [3] where the mapping of luminance
using least mean squared error. To test the performance of these values to color values is automatic, the choice of color map is
algorithms, color image is converted into gray scale image and the
same grayscale image is recolored back. Finally MSE of recolored
commonly determined by human decision. The main concept of
image and original image is compared. Experiment is conducted colorization techniques exploits textual information. The work
on both RGB and Kekre’s LUV color space for the different pixel of Welsh et al , which is inspired by the color transfer [4] and
windows of size 1x2, 2x1, 2x2, 2x3, 3x2, 3x3, 1x3, 3x1, 2x4, 4x2, by image analogies [5], examines the luminance values in the
1x4, 4x1. However Kekre’s LUV color space gives outstanding neighborhood of each pixel in the target image and add to its
performance. For different pixel windows KPE with 1x2 and LBG luminance the chromatic information of a pixel from a source
with 2x1 pixel window perform well with respect to image quality. image with best neighborhoods matching .This technique works
on images were differently colored regions give rise to distinct
Keywords- Colorization , Pixel Window, ColorPalette, Vector
textures otherwise, the user must specify rectangular swatches
Quantization(VQ) , LBG, KPE.
indicating corresponding regions in the two images.
Color traits transferred to gray scale images [6] presents novel
I. INTRODUCTION coloring techniques where color palette is prepared using pixel
Colors always provide more clear information than gray windows of some degree taken from reference coloring image.
scale digital images. Colorization is the art of adding color to a For every window of gray scale image the palette is searched
monochrome image or movie. Colors we perceive in an object for equivalent color values which could be used to color gray
are determined by nature of light reflected from the object. Due scale window [19].
to the structure of human eye, all colors are seen as variable
combinations three basic colors Red, Green, Blue (RGB). The In this paper, adjacent pixels are grouped together to form a
task of coloring a grayscale image involves assigning RGB (pixel window) grid. Vector Quantization algorithms LBG and
values to an image which varies along only the luminance KPE are applied on different pixel window sizes 1x2, 2x1, 2x2,
value. Since different colors may have the same luminance but 2x3, 3x2, 3x3, 1x3, 3x1, 2x4, 4x2, 1x4, 4x1and codebook of size
vary in hue and saturation, the problem of coloring gray scale 512 is obtained. Vector Quantization algorithms LBG and KPE
needs human interaction [1]. are applied. Depending on minimum Euclidean distance, LUV

134 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 6 June 2011
components of reference image are transferred to input gray two vectors are generated by using constant error addition to the
image. codevector. Euclidean distances of all the training vectors are
computed with vectors v1 & v2 and two clusters are formed
II. KEKRE’S LUV COLOR SPACE [15,21] based on closest of v1 or v2. This modus operandi is replaced
for every cluster. The shortcoming of this algorithm is that the
In the proposed technique Kekre’s LUV color space is used. cluster elongation is +135O to horizontal axis in two dimensional
Where L gives luminance and U and V gives chromaticity cases resulting in inefficient clustering.
values of color image. Positive values of U indicate prominence
of red components in color image and negative value of V
indicates prominence of green component. The RGB-to LUV
and LUV-to-RGB conversion matrices are given in equation 1
and 2 respectively.

⎡L ⎤ ⎡ 1 1 1⎤ ⎡R ⎤
⎢U ⎥ = ⎢− 2 1 1⎥ * ⎢G ⎥ (1)
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢V ⎥
⎣ ⎦ ⎢ 0 1 1⎥
⎣ ⎦ ⎢B ⎥
⎣ ⎦
⎡ R ⎤ ⎡1 − 2 0⎤ ⎡ L / 3 ⎤ Figure1 LBG for Two dimensional case.
⎢G ⎥ = ⎢1 1 1 ⎥ * ⎢U / 6⎥ (2)
B. Kekre’s Proportionate Error (KPE) Algorithm [9,10]
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ B ⎥ ⎢1 1 1 ⎥ ⎢V / 2 ⎥ Here to generate two vectors v1 & v2 proportionate error is
⎣ ⎦ ⎣ ⎦ ⎣ ⎦ added to the codevector. Magnitude of elements of the
codevector decides the error ratio. Hereafter the procedure is
same as that of LBG. While adding proportionate error a safe
III. VECTOR QUANTIZATION guard is also introduced so that neither v1 nor v2 go beyond the
training vector space eliminating the disadvantage of the LBG.
Vector Quantization (VQ) [7],[8] is an efficient and lossy Fig. 2, shows the cluster elongation after adding proportionate
technique for compression of data and has been successfully error.
used in various applications like an pattern recognition[11],
speech recognition and face detection[12][13],image
segmentation[14],speech data compression [16],content based
image retrieval CBIR[17],[18] etc.

Vector Quantization can be define as a mapping function
that maps k-dimensional vector space to a finite set CB = {C1,
C2,C3, ..…., CN}. The set CB is called codebook consisting of
N number of codevectors and each codevector Ci= {ci1, ci2, ci3,
……, cik} is of dimension k. The key to VQ is the good
codebook. Codebook can be generated in spatial domain by Figure 2 orientation of line joining two vectors v1 and v2 after addition of
clustering algorithms. proportionate error to the centroid.

IV. PROPOSED COLORING TECHNIQUE
In color transfer phase, image is divided into non
overlapping blocks and each block then is converted to the Since the coloring problem always requires human
training vector Xi = (xi1, xi2, ……., xik ). The codebook is then interaction. So reference image of same class and of same
searched for the nearest codevector Cmin by computing squared feature as of input grayscale image. The color transfer
Euclidian distance as presented in equation (3) with vector Xi algorithm is discussed for LUV color space for different m x n
with all the codevectors of the codebook CB. This method is pixel grid size. The main steps of algorithm for a color transfer
called exhaustive search (ES). are:
d(Xi, Cmin) = min1≤j≤N{d(Xi,Cj)} (3) • Convert RGB components of source color image into
where d(Xi,Cj) = ∑(Xip - Cjp)2 respective Kekre’s LUV color components.
It is obvious that, if the codebook size is increased to reduce the • Divide the image in to blocks of m x n pixels. Hence
distortion the searching time will also increase. m x n x3 dimensional training vector set
The following section describes the VQ codebook corresponding to LUV components of each pixel is
Generation Algorithms. obtained. On this set LBG and KPE algorithms are
applied and color palette is generated i.e. codebook of
A. Linde Buzoand Gray Algorithms(LBG) [7,8] size 512.
• The input gray image is divided in mxn blocks of
In this algorithm centroid is first calculated by taking
pixels. Each block (pixel window) is searched for
average as the first code vector for the training set. In figure1

135 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 6 June 2011
• nearest code vector of color palette. While searching
only luminance is compared
• Once the nearest match is obtained gray image pixel
window is replaced by LUV codevector
• The final colored image in LUV domain is then
converted into RGB plane and MSE of original color
image and recolored image are calculated.
6(a) 6(b) 6(c) 6(d)
Original image Gray image
V. RESULTS 1x2 Grid LBG 1x2 Grid KPE
MSE: 92 MSE: 81
The algorithms discussed above are implemented using Figure 6 shows reconstruction of face grayscale image using similar source
image for pixel window 1x2.
MATLAB 7.0 on Pentium IV, 1.66GHz, 1GB RAM. To test the
performance of these algorithms we have converted color
image to grayscale image and the same gray image is recolored
back. Finally MSE of original image and colored image is
compared. Five color images belonging to different classes of
size 128x128x3 are used. 7(a) 7(b) 7(c) 7(d) 7(d)
Figure3 to Figure6. Shows the results of LBG and KPE for Original Reference Gray 1x2 Grid 1x2 Grid KPE
Zebra, Book, Cartoon and Face images considering same image Image Image Image LBG MSE 709
MSE 990
as reference image. Figure 7 shows reconstruction of Scenery grayscale image using
Figure7 and Figure8. Shows the results of LBG and KPE for different source image.
scenery and dog images considering different image as
reference image.

8(a) 8(b) 8(c) 8(d) 8(e)
Original Reference Gray Image 1x2Grid 1x2 Grid
Image Image LBG KPE MSE
MSE 303 285
Figure 8 shows reconstruction of Dog grayscale image using different
source image.
3(a) 3(b) 3(c) 3(d)
Original Gray image 1x2 Grid LBG 1x2 Grid KPE Various images, each of size 128x128 pixels, were
image MSE: 178.4 MSE: 73.85 used to build the color palette, and their grayscale
Figure 3 shows reconstruction of Zebra grayscale image using similar equivalents were colored using color palette for various
source image for pixel window 1x2
pixel windows. The fig. 9, shows bar chart of average
mean squared error obtained across all five images with
respect to initial few pixel windows for RGB and Kekre’s
LUV color space. It is observed that, Kekre’s LUV color
space gives less MSE compared to RGB color space.
Hence in table 1 only Kekre’s LUV color space results for
different images using 12 varying pixel window
sizes(1x2,2x1,2x2,2x3,3x2,3x3,1x3,3x1,2x4,4x2,1x4,4x1)
4 (a) 4(b) 4(c) 4(d) are given.
Original Image Gray Image 1x2GridLBG 1x2 Grid KPE
MSE73.8 MS53.32
Figure 4 shows reconstruction of book grayscale image using similar
source mage for pixel window 1x2

Figure 9– Average MSE across various Grid sizes for different color
spaces
5(a) 5(b) 5(c) 5(d)
Original image Gray image 1x2 Grid LBG 1x2 Grid KPE
MSE: 1260 MSE:1023
Figure 5 shows reconstruction of cartoon grayscale image using similar
source mage for pixel window 1x2

136 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 6 June 2011

Table I. Shows the Results of LBG and KPE for five color images from different categories of size 128x128x3.

Input VQ Grid Sizes
Images Alg.
1x2 2x1 2x2 2x3 3x2 3x3 2x4 4x2 1x3 3x1 1x4 4x1
Image1 LBG 92.70 87.11 107.42 116.75 114.65 2302 5576 5652 107.32 106 122.29 114.
KPE 81.59 77.49 89.141 90.75 95.83 2122 5557 5663 87.32 89.09 91.41 87.3
Image2 LBG 1260 1244 1056 1493 1420 6350 6928 9595 1251 1243 1532 1392
KPE 1023 1153 1363 1233 1150 6243 6725 9359 1278 1138 1013 1093
Image3 LBG 178.4 147 440.90 721.73 675.15 2003 2486 4236 373 286 610.66 465.7
KPE 73.85 76.12 225.10 461.69 441.62 1823 2264 4663 143.5 142 273.38 226.0
Image4 LBG 73.89 76.64 107.12 123.82 131.86 2340 3400 3664 90 97 116.92 128.9
KPE 53.32 52.73 79.094 103.43 95.439 2813 3371 3701 65.3 65 73.93 76.09
Image5 LBG 1203 1244 1244 1406 1388 6340 7833 7916 1246 1240 1274 1266
KPE 1178 1174 1182 1414 1399 6384 7935 7968 1193 1175 1209 1212
Average LBG 561.5 559.7 591 772.26 745.9 3867 5244 6212 613 594 731 673
Average KPE 481.9 506.6 587.8 660.5 636.3 3877 5170 6270 553.4 521.8 532.1 538.8

From the data given in table1, it is seen that the performance REFERENCES
gradually decreases as the pixel window size increases.Further [1] V. Karthikeyani, K. Duraisamy, Mr.P.Kamalkakkannan, " Conversion
MSE for unidirectional pixel window is less compared to of grayscale image to color image with and without texture synthesis",
IJCSNS International journal of Computer science and network
bidirectional. Pixel window sizes 1x2 and 2x1 are showing security, Vol.7 No.4 April 2007.
better results as compared to large pixel window sizes. [2] E.Reinhard, M. Ashikhmin, B. Gooch and P Shirley, “Colour Transfer
Fig.10, shows the comparison of average mean sqared error between images”, IEEE Transactions on Computer Graphics and
obtained across all images on Kekre’s LUV color space for top Applications 21, 5, pp. 34-41.
five pixel windows. It can be seen from the chart , KPE [3] Rafael C. Gonzalez & Paul Wintz, “ Digital Image Processing”,
Addison Wesley Publications, May 1987.
performs well with respect to LBG. Also performance
[4] A. Hertzmann, C. E Jacobs, N. Oliver, B. Curless and D.H. Salesin,
deteriorates as pixel window size increases and becomes “image Anologies”, in the proceedings of ACM SIGGRAPH 2002, pp.
bidirectinal. 341-346.
[5] G. Di Blassi, and R. D. Reforgiato, “Fast colourization of gray
images”, In proceedings of Eurographics Italian Chapte, 2003.
[6] H.B.Kekre, Sudeep. D. Thepade, “Color traits transfer to gray scale
images”, in Proc of IEEE International conference on Emerging Trends
in Engineering and Technology, ICETET 2008 Raisoni College of
Engg, Nagpur.
[7] R. M. Gray, "Vector quantization", IEEE ASSP Mag., pp. 4-29,
Apr11984.
[8] Y. Linde, A. Buzo, and R. M. Gray, "An algorithm for vector quantizer
design," IEEE Trans.Commun., vol. COM-28, no. 1, pp. 8495, 1980.
Figure 10 Average MSE across various Grid sizes
[9] H. B. Kekre, Tanuja K. Sarode, "New Fast Improved Codebook
VI. CONCLUSION Generation Algorithm for Color Images using Vector Quantization,"
International Journal of Engineering and Technology, vol.1, No.1, pp.
In this paper, the idea of colorization of grayscale images 67-77, September 2008.
using VQ codebook generation techniques is presented using [10] H. B. Kekre, Tanuja K. Sarode, "An Efficient Fast Algorithm to
two famous codebook generation algorithms alias LBG and Generate Codebook for Vector Quantization," First International
KPE. For both the algorithms 12 assorted pixel window sizes are Conference on Emerging Trends in Engineering and Technology,
ICETET-2008, held at Raisoni College of Engineering, Nagpur, India,
considered for preparing the color palettes. As quality of July 2008, Available at online IEEE Xplore.
colorization is subjective to source color image and grayscale to [11] Ahmed A. Abdelwahab, Nora S. Muharram, "A Fast Codebook Design
be colorized image, the grayscale version of 5 color images are Algorithm Based on a Fuzzy Clustering Methodology", International
recolored using total 48 variations of proposed techniques with 2 Journal of Image and Graphics, vol. 7, no. 2 pp. 291302, 2007.
color spaces (RGB and Kekre’s LUV), 12 pixel window sizes [12] Chin-Chen Chang, Wen-Chuan Wu, "Fast Planar-Oriented Ripple
and 2 codebook generation techniques( LBG and KPE). The Search Algorithm for Hyperspace VQ Codebook", IEEE Transaction
comparison of original color image and recolored image has on image processing, vol 16, no. 6, June 2007.
shown that Kekre’s LUV color space outperforms RGB color [13] C. Garcia and G. Tziritas, "Face detection using quantized skin color
space. Further, it can be observed from results that unidirectional regions merging and wavelet packet analysis," IEEE Trans.
Multimedia, vol. 1, no. 3, pp. 264-277, Sep. 1999.
pixel windows gives better colorization than bidirectional pixel
[14] H. B. Kekre, Tanuja K. Sarode, Bhakti Raul, "Color Image
window sizes. The KPE performs better than LBG for Segmentation using Kekre's Fast Codebook Generation Algorithm
colorization. In all the best performance is shown by KPE with Based on Energy Ordering Concept", ACM International Conference
1x2 window size in Kekre’s LUV color space. on Advances in Computing, Communication and Control (ICAC3-
2009), 23-24 Jan 2009, Fr. Conceicao Rodrigous College of Engg.,
Mumbai. Available on online ACM portal.

137 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 9, No. 6 June 2011
[15] Dr.H.B. Krkre, Sudeep D. Thepade, “Image Blending in Vista Creation Dr. Tanuja K. Sarode has Received Bsc.(Mathematics) from Mumbai
using Kekre’s LUV Color Space”, In Proc. Off PIT-IEEE Colloquium, University in 1996, Bsc.Tech.(Computer
Mumbai, Feb 4-5,2008. Technology) from Mumbai University in 1999,
[16] H. B. Kekre, Tanuja K. Sarode, "Speech Data Compression using M.E. (Computer Engineering) degree from Mumbai
Vector Quantization", WASET International Journal of Computer and University in 2004, Ph.D. from Mukesh Patel
Information Science and Engineering (IJCISE), vol. 2, No. 4, 251254, School of Technology, Management and
Fall 2008. available: http://www.waset.org/ijcise. Engineering, SVKM’s NMIMS University, Vile-
[17] H. B. Kekre, Ms. Tanuja K. Sarode, Sudeep D. Thepade, "Image Parle (W), Mumbai, INDIA. She has more than 12
Retrieval using Color-Texture Features from DCT on VQ Codevectors years of experience in teaching. Currently working
obtained by Kekre's Fast Codebook Generation", ICGST-International as Assistant Professor in Dept. of Computer
Journal on Graphics, Vision and Image Processing (GVIP),Volume 9, Engineering at Thadomal Shahani Engineering
Issue 5, pp.: 1-8, September 2009. Available online at College, Mumbai. Engineering, SVKM’s NMIMS University, Vile-Parle (W),
http://www.icgst.com/gvip/Volume9/Issue5/P1150921752.html. Mumbai, INDIA. She has more than 12 years of experience in teaching.
Currently working as Assistant Professor in Dept. of Computer Engineering at
[18] H.B.Kekre, Tanuja K. Sarode, Sudeep D. Thepade, "Color-Texture
Thadomal Shahani Engineering College, Mumbai. She is life member of IETE,
Feature based Image Retrieval using DCT applied on Kekre's Median
member of International Association of Engineers (IAENG) and International
Codebook", International Journal on Imaging (IJI),Available online at
Association of Computer Science and Information Technology (IACSIT),
www.ceser.res.in/iji.html.
Singapore. Her areas of interest are Image Processing, Signal Processing and
[19] Dr. H. B. Kekre, Sudeep D. Thepade, Nikita Bhandari, “Colorization of Computer Graphics. She has 90 papers in National /International
Gereyscale images using Kekre’s Bioorthogonal Color Spaces and Conferences/journal to her credit.
Kekre’s Fast Codebook Generation “,CSC Advances in Multimedia
An international journal (AMU), volume 1, Issue 3,pp.49-58, Available Sudeep D. Thepade has Received B.E.(Computer) degree from North
at Maharashtra University with Distinction in
www.cscjournals.org/csc/manuscript/journals/AMIJ/volume1/Issue3/A 2003. M.E. in Computer Engineering from
MU-13.pdf. University of Mumbai in 2008 with
[20] Dr. H. B. Kekre, Sudeep D. Thepade,Adib Parkar, “A Comparison of Distinction, currently submitted thesis for
Harr Wavelets and Kekre’s Wavelets for Storing Color Information in Ph.D. at SVKM’s NMIMS, Mumbai. He has
a Greyscale Images”, International Journal of Computer more than 08 years of experience in
Applications(IJCA), Volume 1, Number 11, December 2010,pp 32-38. teaching and industry. He was Lecturer in
Available at www.ijcaonline.org/archives/volume11/number11/1625- Dept. of Information Technology at
2186. Thadomal Shahani Engineering College,
[21] Dr. H. B. Kekre, Sudeep D. Thepade,Archana Athawale, Adib Parkar, Bandra(w), Mumbai for nearly 04 years.
“Using Assorted Color Spaces and pixel window sizes for Colorization Currently working as Associate Professor in
of Grayscale images’,ACM International Conferences and workshops Computer Engineering at Mukesh Patel School of Technology Management
on emerging Trends in Technology(ICWET 2010), Thakur College of and Engineering, SVKM’s NMIMS, Vile Parle(w), Mumbai, INDIA. He is
Engg. And Tech.,Mumbai,26-27 Feb 2010. member of International Association of Engineers (IAENG) and International
[22] H. B. Krekre,Sudeep Thepade, Adib Parkar, “A comparison of Kekre’s Association of Computer Science and Information Technology (IACSIT),
Fast Search and Exhaustive Search for various grid sizes used for Singapore. He is member of International Advisory Committee for many
coloring a Grayscale Image” Second International conference on signal International Conferences. He is reviewer for various International Journals.
Acquisition and Processing, (ICSAP2010), IACSIT,Banglore,pp.53- His areas of interest are Image Processing Applications, Biometric
57,9-10 Feb 2010. Identification. He has about 110 papers in National/International
Conferences/Journals to his credit with a Best Paper Award at International
Conference SSPCCIN-2008, Second Best Paper Award at ThinkQuest-2009
Author Biographies National Level paper presentation competition for faculty, Best paper award at
Springer international conference ICCCT-2010 and second best research project
Dr. H. B. Kekre has received B.E. (Hons.) in Telecomm. Engineering. from award at ‘Manshodhan-2010’.
Jabalpur University in 1958, M.Tech
(Industrial Electronics) from IIT Bombay in
1960, M.S.Engg. (Electrical Engg.) from Supriya Kamoji has received B.E. in Electronics and Communication
University of Ottawa in 1965 and Ph.D. Engineering with Distinction from Karnataka
(System Identification) from IIT Bombay University in 2001. Currently pursuing M.E. from
in 1970 He has worked as Faculty of Thadomal Shahani College of Engineering,
Electrical Engg. and then HOD Computer Mumbai, India. She has more than 8years of
Science and Engg. at IIT Bombay. For 13 teaching experience. Currently working as an
years he was working as a professor and head Senior Lecturer in Fr.Conceicao Rodrigues
in the Department of Computer Engg. at College of Engineering. Mumbai, India. She is a
Thadomal Shahani Engineering. College, Mumbai. Now he is Senior Professor life time member of Indian society of Technical
at MPSTME, SVKM’s NMIMS. He has guided 17 Ph.Ds, more than 100 Education (ISTE). Her areas of interest are Image
M.E./M.Tech and several B.E./ B.Tech projects. His areas of interest are Digital Processing, Computer Organization and Architecture and Distributed
Signal processing, Image Processing and Computer Networking. He has more Computing.
than 270 papers in National / International Conferences and Journals to his
credit. He was Senior Member of IEEE. Presently He is Fellow of IETE and
Life Member of ISTE Recently 11 students working under his guidance have
received best paper awards. Two of his students have been awarded Ph. D. from
NMIMS University. Currently he is guiding ten Ph.D. students.

138 http://sites.google.com/site/ijcsis/
ISSN 1947-5500