|
Evolution of Semantic Schema Matching as a Web Service
-- S Vasavi and L S S Reddy
The term `interoperability' means using data and services that are defined independent of the
application, programming language and operating system. Initially, interoperability has been achieved by setting standards
on either ends of exchange parties such as Electronic Data Interchange (EDI), The Workflow Management
Coalition (WfMC), eCo, Electronic Business using XML (ebXML), cXML, Business Process Modeling Notation (BPMN),
Universal Business Language (UBL), RosettaNet, Society for Worldwide Interbank Financial Telecommunication (SWIFT)
and STandard for the Exchange of Product Model Data (STEP). These standards enabled interoperability without
human intervention, but forced all applications to use a standard structured format. Later, object models such as
CORBA, COM, DCOM, COM+, EJB, JVM were introduced to solve the interoperability problem. But these technologies
were complex and tightly coupled, client server-based architectures. Even though web services are loosely coupled,
they require a new layer above the existing web service stack for resolving mismatches during interoperability of data
or services or both. This paper describes such layer as a service called semantic schema matching which is
initially designed for performing schema matching over diverse data sources and later exposed as a web service. A road
map for evolving this service for use by public clients is also discussed.
© 2009 IUP. All Rights Reserved.
Three-Pass Cryptosystems Based on Discrete Logarithms
-- P S Gill and Ashish Kr. Srivastava
In cryptography, a `Three-Pass Protocol' facilitates a secure communication of confidential messages, over
insecure channels, without the need of any exchange of keys. Each communicating entity is required to generate a pair of
keys, related to each other. One of the keys is used for encryption and the other key is used for decryption. Transmission of
information between a sender and the intended recipient requires making of three passes. Each pass involves
exchange of an encrypted message between the communicating entities. In pass 1, the sender encrypts the plain text with one
of its keys and sends the resulting cipher text to the intended recipient. The intended recipient further encrypts
the received cipher text with one of its keys and bounces the doubly-encrypted cipher text back to the sender. This
cipher text exchanged in pass 2 has double encryption—one applied by the sender and the other applied by the
intended recipient. The sender decrypts the doubly-encrypted cipher text using its second key and removes its part of
the encryption. The resulting cipher text now has only one encryption—the one applied by the intended recipient. In
pass 3, this singly-encrypted cipher text is sent by the sender to the intended recipient. The intended recipient receives
the cipher text and removes the residual encryption using its second key; and successfully recovers the original plain
text, meant to be conveyed to the intended recipient in a secure way. In all the three passes, the message is
encrypted—having single encryption during passes 1 and 3 and double encryption during pass 2. Thus, the two
communicating entities are able to exchange information in a secure way, without any need of exchange of keys.
© 2009 IUP. All Rights Reserved.
Underlying Dynamics of Carriers in the Study of Epidemic Models
-- Nistala Suresh Rao, Devanand and
Peri Sarveswara Avadhani
In this paper, the growth of computer virus, in a given population of computers, based on
Susceptible-Infected-Susceptible (SIS) model is investigated taking into account the role of
Carriers. Three situations are considered:
Firstly, Carriers remain constant and they only spread the virus.
Secondly, Carriers remain constant and the virus is spread due
to infectives and Carriers. Thirdly, the number of
Carriers decreases with time. In all the three
cases,
the asymptotic behavior of the number of infected computers with respect to time is investigated. In this
analysis, the important parameters are: population size, epidemic threshold, birth and death rates of virus and the number of
Carriers. In the first two cases, the growth/fall of the number of infected computers asymptotically reaches a saturation value
and remains constant for large values of time. In the third
case, the same increases and reaches a maximum value and
then asymptotically falls to zero.
© 2009 IUP. All Rights Reserved.
Video-Based Person Authentication Using Face
and Visual Speech
-- M Balasubramanian, S Palanivel and V Ramalingam
This paper proposes a facial and visual speech feature extraction method for automatic person authentication in video.
The method proposed in Viola and Jones (2006) is used to detect the face region. Face region is processed in
YCbCr color space to determine the locations of the eyes. The system models the non-lip region of the face using a Gaussian distribution,
and it is used to locate the center of the mouth. Facial and visual speech features are extracted using multiscale
morphological erosion and dilation operations, respectively. The facial features are extracted relative to the locations of the eyes and
visual speech features are extracted relative to the locations of the eyes and mouth. Auto-Associative Neural Network (AANN)
and Support Vector Machines (SVMs) are analyzed for person authentication. AANN models are used to capture the
distribution of facial and visual speech features of a subject. SVMs are used to construct the optimal separating hyperplane for facial
and visual speech features. The evidence from face and visual speech modalities are combined using a weighting rule, and
the result is used to accept or reject the identity claim of the subject. The performance of the system is evaluated for
XM2VTS database. It is seen that the system achieves an Equal Error Rate (EER) of about 0.41% and 0.37% for 50 subjects using
AANN and SVM, respectively. Finally, the performance of the AANN and SVM models for person authentication are
compared. Experimental results show that the SVM gives better performance than the AANN model.
© 2009 IUP. All Rights Reserved.
Querying the Distributed Multicolor Database
-- R Seethalakshmi
Querying the object databases is quite an important aspect when dealing with a variety of database objects. A
multicolor database contains objects that are composed of
spectral colors. In such multicolor database context, it is very
important to recognize particular object or objects, by querying the database. The query can contain options like retrieving
objects by color, retrieving objects by providing texture, retrieving objects by histogram. In retrieving the objects by color,
the query takes the color intensity value and the database is processed to check the objects for the respective maxcolor
by computing the histogram of the objects. These objects, which match the specified color,
and are retrieved. The retrieval provides tolerance by a threshold. In the context of retrieving the objects by texture, the sample texture is chosen
from query database and given as a query. The texture is moved over the entire object and if there is a match,
satisfying the specified threshold, those objects will be chosen as retrieved. In this
paper, we also address the issue of retrieving
the object by querying database for a particular histogram. The histograms are computed dynamically in
runtime. Those which, match the query histogram,
are retrieved as the result of the query. Sophisticated algorithms and databases
are designed and implemented in a distributed way. The various databases like query database, object database,
histogram database, color palette databases are created, fragmented and distributed, across the network.
Thus, there is a retrieval of multicolored objects in
a faster and efficient manner.
© 2009 IUP. All Rights Reserved.
Model for Color Analysis
-- Dipti Shah
In image analysis, basic objects and their relationships are extracted from a picture given in unstructured form.
Color is one of the powerful descriptors of an image. The color, we see, depends on the type of graphic file format,
resolution of an image, and resolution and type of input/output devices. Also, the computer identifies color by its Red, Green
and Blue components value and human eyes identify color by its name. Moreover, human eye has several limitations
in processing the color like it cannot distinguish between all colors and cannot resolve the components of a color,
i.e., visually one cannot find and input values of individual R, G and B components of a particular color. The model
for color analysis is designed and developed to filter patterns of colors from the whole image or interested Region
of Image (ROI), as per the control parameter. The control parameter could be color value, one point of an image,
color name or range of colors. Depending on the value of the control parameter, the model filters the color information
from the image. The model helps the domain experts of diverse area to study the various relationships among the
different parameters in an image.
© 2009 IUP. All Rights Reserved.
Research Note
A Note on Advancement of Tree
"One could...take evolutionary bibliography as the
prototypical evolutionary science and think of
biology in terms of bibliographic analogies..."
-- Sanjay Kumar Pal, Samar Sen Sarma and Anindya Jyoti Pal
© 2009 IUP. All Rights Reserved.
|
