P. Karmuhil ¹, Latha Parthiban ²

Affiliations
1 Periyar University, Salem - 636011, Tamil Nadu, IN
2 Pondicherry University Community College, Puducherry - 605008, IN

Abstract

Background/Objectives: The objective of this study is detection and reconstruction of shadows from Very High Spatial Resolution Images (VHSR images). Methods/Statistical Analysis: This paper, presents an effective unsupervised method of segmentation for detecting shadows using modified Self Organizing Maps (SOM and MRF). This term MRF is adapted with SOM to segment the shadows without giving more trained sample data set and linear regression method is applied in shadow regions for compensating the shadows by non shadow regions (Reconstruction process). Findings: Experimental result of proposed method gives a significant result on VHSR images and assess the various quality metrics to obtain the better performance, quality and accuracy than the other conventional methods. Application/Improvement: The Modified SOM network eliminates the fake shadows, avoids over segmentation reduces the execution time for segmenting the shadows and improves accuracy as appx 92%. It is mainly used for various remote sensing applications such as change identification, object recognition, scene restoration, color tuning etc.

Keywords

Linear Regression Method, MRF, Reconstruction, Self-organizing Maps, Shadow Detection, VHSR images

Full Text

   

B. Santhosh Kumar ¹, Latha Parthiban ²

Affiliations
1 Department of CSE, G. Pulla Reddy Engineering College, Kurnool – 518007, Andhra Pradesh, IN
2 Department of CSE, Community College, Pondicherry University, Pondicherry – 605014, Tamil Nadu, IN

Abstract

Background: Data centers are the major sources for implementing critical issues in a cloud computing environment like networking, task scheduling, storage and the distribution of data. The objective is to design a data centre selection framework for submission of tasks to a data centre with minimal energy consumption. Methods: A framework is proposed to select a data centre for tasks submission with minimum total energy consumed by the servers, computer room air conditioning units and other IT equipments like routers, switches. The service provider has to register all the data centers in a data centre registry and the data centre selection framework decides upon task allocation to the appropriate data centre with minimal energy consumption. The results are simulated using Google Cloud Trace logs and cloud analyst tool. The proposed work is compared with shortest distance first and round robin methods. Findings: The performance analysis indicate that the proposed framework is an efficient way for tasks submission to a data centre as it concentrates on minimizing the energy consumption which in turn reduces operational expenditures of the cloud environment. The comparisons of the work with other algorithms indicate that the proposed framework consumes far less energy. Improvements: The framework can be implemented by tools like IBM Bluemix which uses Internet of Things (IoT) technology for reading the energy using sensors.

Keywords

Cloud Computing Environment, Data Centre Selection Framework, Energy Consumption, Tasks Submission

Full Text

   

Sabout Nagaraju ¹, Latha Parthiban ¹

Affiliations
1 Department of Computer Science, Pondicherry University Community College, Lawspet, Pondicherry-605 008, IN

Abstract

Cloud computing is an advanced resource pooling framework which delivers an economical and more reliable Information and Communications Technology (ICT) solutions to the industry and academia. Cloud technology helps the stakeholders to avoid initial investment of costly infrastructure setup, licensing new software’s, training new personals and operational cost. Therefore, all size of IT organizations and individuals can make use of the cloud for boosting up their ICT needs. In parallel there has been backlash due to authentication access keys and credentials management issues in this new framework. As a result, in this article we have proposed a robust and privacy preserved Multi-Factor Authentication (MFA) scheme with efficient access keys distribution mechanism. The proposed MFA approach integrates the bio-metric fingerprint with user-id, password and One-Time Password (OTP) and upgrades the existing Single Sign-On (SSO) and two-factor authentications to multi-factor authentication. Our investigation not only provides the robust remote authentication in cloud but also preserves the privacy of the authentication credentials. The main shortcoming in our approach is that remote users and cloud service providers must trust the third party trustee. We have analyzed the completeness of our investigation using the GNY (Gong, Needham and Yahalom) logic. Finally, we reported the performance and robustness of our scheme with series of experiments.

Keywords

Authentication, Cloud Computing, Fingerprint, Privacy, Security, Trustee

Full Text

   

T. P. Latchoumi ¹, Latha Parthiban ²

Affiliations
1 Department of Computer Science and Engineering, Sathyabama University, Chennai - 600119, Tamil Nadu, IN
2 Department of Computer Science and Engineering, Community College, Pondicherry University, Pondicherry – 605014, Tamil Nadu, IN

Abstract

Objectives: The objective is to introduce multi-agent systems to enhance the security rules through the access right to build a distributed warehouse in the cloud environment secure manner. The work aim is to take a smart decision making using MAS in a timely manner. Analysis: Cloud computing is a very powerful, and predictable in computing infrastructure for implementing complex agent-based applications. Cloud computing in Multi-Agent System (MAS) appears as an approach to current challenges in many areas. The distributed data warehousing is used to do based on how data’s are distributed in the multiple servers.Its main problemishow the Collaborative work of a multi-agent system designed for distributed data warehousing in the cloud environment. Findings: In the existing approach, a major issue is to retrieve the relevant data from the cloud storage is a very difficult task. Other issues in attributes are increasing the network loads (traffics) and response time, the need of secure data storage, data retrieval from the cloud environment and the database updating is very slow.Cloud infrastructures provide a platform to run the MAS in the real-time, because it takes large execution time by havinga large amount of data processing and dynamic memory. Improvement: In this paper, we introduce data warehouse in the cloud computing through the multi-agent system technology. This enables cost and time saving. The technique based on data warehouse in the cloud environment using Multi-Agent Systems (MAS) technology to consider security and privacy in data storage and transmitted. The proposed system is the use of an MAS in the cloud environment, introduces autonomous decision making in the critical situation to speed up the execution time, response time, database updating and security enhancing. We apply this propose system in any application like e-banking, hospital management, election department, etc.

Keywords

Cloud Computing, Distributed Data Warehousing, Multi Agent Systems (MAS), Query Redirection Process, Security

Full Text