Ophthalmic diseases like diabetic retinopathy,
macular degeneration, glaucoma, etc. may cause gradual loss
of eyesight and are some of the reasons behind blindness.
Hence, blood vessel assessment and segmentation play a key
role in the diagnosis of retinal disorders. The manual
detection of narrow blood vessels of retinal images is time
consuming and may result in erroneous output. Therefore,
computer aided, robust and performance oriented algorithm
is required to detect and segment the blood vessels efficiently.
The present work proposes an algorithm that uses
thresholding technique, basic morphological operations and
Kirsch’s edge detection operator to detect the blood vessels
and segment the hard exudates efficiently. The detected
exudates regions are then compared with the ground truth
exudate regions. Based on the correctly identified exudate
regions between these two images, different performance
parameters like accuracy, specificity, sensitivity, PPV, PLR
and misclassified proportions have been measured to
evaluate overall performance of the proposed algorithm. This
algorithm successfully detects exudates with an average of
98.47% accuracy, 54.67% sensitivity, 99.82% specificity,
88.62% PPV, 303.73 PLR and 0.17% misclassified
proportions.
Published In:IJCSN Journal Volume 5, Issue 4
Date of Publication : August 2016
Pages : 615-621
Figures :03
Tables : 02
Shreyasi Hazra : received her B. Tech degree in Electronics and
Communication Engineering from JIS College of Engineering
under West Bengal University of Technology in 2012. She has
worked with IBM as Application Developer for 1 year and 8
months. Currently, she is pursuing her M.Tech in Electronics and
Communication from Institute of Engineering & Management
under same university. Her area of interest includes image
processing & analysis.
Atashi Patra : has completed her B. Tech in Electronics and
Communication Engineering from Asansol Engineering College
under West Bengal University of Technology in 2013. Currently,
she is pursuing her M.Tech in Electronics and Communication
from Institute of Engineering & Management under same
university. Her area of interest is in image processing & analysis.
Tuhin Utsab Paul : did his undergraduate and postgraduate
degrees in computer science and engineering from the University
of Calcutta. He is working as an assistant professor in the
department of electronics and communication at the Institute of
Engineering and Management since 2012. His research area
includes bio – medical image processing, digital signal processing
and embedded system design.
This paper presents various existing methods from
different papers used for the privacy preserving query
processing in data mining and over encrypted data can
mention as (PPDM). To protect user data privacy, with
the discussion of various proposals of privacy-preserving
classification techniques are presented over the past
papers. The existing techniques are applicable to
outsourced database environments where the data resides
in encrypted form on a third-party server and also the
classification or clustering of the data with proper label is
discussed here This paper proposed a new privacypreserving
clustering protocol DBSCAN is used over
encrypted data of the cloud. This protocol protects the
confidentiality of the data, as a user’s input query, and
hides the data access patterns. As well as provide fast
access to data as better clustering is done using DBSCAN
algorithm.
[1] M. Kantarcioglu and C. Clifton, “Privately computing
a distributed k-nn classifier,” in PKDD, pp. 279–290,
2004.
[2] S. De Capitani di Vimercati, S. Foresti, and P.
Samarati, “Managing and accessing data in the cloud:
Privacy risks and approaches,” in CRiSIS, pp. 1 –9,
2012.
[3] P. Williams, R. Sion, and B. Carbunar, “Building
castles out of mud: practical access pattern privacy and
correctness on untrusted storage,” in ACM CCS, pp.
139–148, 2008.
[4] Y. Qi and M. J. Atallah, “Efficient privacy-preserving
k-nearest neighbor search,” in IEEE ICDCS, pp. 311–
319, 2008.
[5] C. Gentry and S. Halevi, “Implementing gentry’s
fullyhomomorphic encryption scheme,” in
EUROCRYPT, pp. 129–148, Springer, 2011.
[6] A. Shamir, “How to share a secret,” Commun. ACM,
vol. 22, pp. 612–613, Nov. 1979.
[7] D. Bogdanov, S. Laur, and J. Willemson, “Sharemind:
A framework for fast privacy-preserving
computations,” in ESORICS, pp. 192–206, Springer,
2008.
[8] R. Agrawal and R. Srikant, “Privacy-preserving data
mining,” in ACM Sigmod Record, vol. 29, pp. 439–
450, ACM, 2000.
[9] L. Xiong, S. Chitti, and L. Liu, “K nearest neighbor
classification across multiple private databases,” in
CIKM, pp. 840–841, ACM, 2006.
[10] P. Zhang, Y. Tong, S. Tang, and D. Yang, “Privacy
preserving naive bayes classification,” ADMA, pp.
744–752, 2005.
[11] Martin Ester, Hans-Peter Kriegel, Jörg Sander and
Xiaowei Xu, “ A Density-Based Algorithm for
Discovering Clusters in Large Spatial Databases with
Noise”, The Second International Conference on
Knowledge Discovery and Data Mining (KDD-96),
Portland, Oregon, USA, 1996
[12] R. J. Bayardo and R. Agrawal, “Data privacy through
optimal k-anonymization,” in IEEE ICDE, pp. 217–
228, 2005.
[13] H. Hu, J. Xu, C. Ren, and B. Choi, “Processing private
queries over untrusted data cloud through privacy
homomorphism,” in IEEE ICDE, pp. 601–612, 2011.
[14] R. Agrawal, J. Kiernan, R. Srikant, and Y. Xu, “Order
preserving encryption for numeric data,” in ACM
SIGMOD, pp. 563–574, 2004.
[15] B. Hore, S. Mehrotra, M. Canim, and M. Kantarcioglu,
“Secure multidimensional range queries over
outsourced data,” The VLDB Journal, vol. 21, no. 3,
pp. 333–358, 2012.
[16] W. K. Wong, D. W.-l. Cheung, B. Kao, and N.
Mamoulis, “Secure knn computation on encrypted
databases,” in ACM SIGMOD, pp. 139–152, 2009.
[17] X. Xiao, F. Li, and B. Yao, “Secure nearest neighbor
revisited,” in IEEE ICDE, pp. 733–744, 2013.
[18] Y. Elmehdwi, B. K. Samanthula, and W. Jiang,
“Secure k- nearest neighbor query over encrypted data
in outsourced environments,” in IEEE ICDE, pp. 664–
675, 2014.