Global Journal of Computer Science and Technology (GJCST)
Not a member yet
1830 research outputs found
Sort by
Gray Scale and Color Medical Image Compression by Lifting Wavelet; Bandelet and Quincunx Wavelets Transforms : A Comparison Study
Get PDFThe Quincunx wavelet , the lifting Scheme wavelet and the Second generation bandelet transform are a new method to offer an optimal representation for image geometric; we use this transform to study medical image compressed using the Quincunx transform coupled by SPIHT coder. We are interested in compressed medical image, In order to develop the compressed algorithm we compared our results with those obtained by this transforms application in medical image field. We concluded that the results obtained are very satisfactory for medical image domain. Our algorithm provides very important PSNR and MSSIM values for medical images compression
Integrated Biometric Template Security using Random Rectangular Hashing
Get PDFLarge centralized biometric databases, accessible over networks in real time are especially used for identification purposes. Multimodal biometric systems which are more robust and accurate in human identification require multiple templates storage of the same user analogous to individual biometric sources. This may raises concern about their usage and security when these stored templates are compromised since each person is believed to have a unique biometric trait. Unlike passwords, the biometric templates cannot be revoked and switch to another set of uncompromised identifiers when compromised. Therefore, fool-proof techniques satisfying the requirements of diversity, revocability, security and performance are required to protect stored templates such that both the security of the application and the users2019; privacy are not compromised by the impostor attacks. Thus, this paper proposes a template protection scheme coined as random rectangular hashing to strengthen the multimodal biometric system. The performance of the proposed template protection scheme is measured using the fingerprint FVC2004 and PolyU palmprint database
Energy Efficient QoS Routing Protocol based on Genetic Algorithm in MANET
Get PDFAbstract- In mobile ad-hoc networks (MANETs), providing quality of service is more challenging than wired networks, because of multi hop communication, node connectivity and lack of central co-ordination. Mobile ad-hoc networks need sure distinctive characteristics which might cause difficulties providing QoS in such network. Coming up with of multi constrained QoS routing protocols remains troublesome. As a result of routing protocols must satisfy the numerous QoS metrics at a time. Genetic algorithm based routing protocol will give the solution for multi constrained QoS routing problem. In existing genetic algorithm based routing, achieving energy efficiency is the major drawback. To overcome this drawback, in this paper, we have proposed genetic algorithm based energy efficient QoS routing for MANET. Proposed GA based routing algorithm discovered the shortest path from source to destination, which can consumes less energy compare to existing algorithms. In this paper TCP,CBR and video sources are applied at a time then energy consumption of proposed algorithm is compared with existing normal GA based and AOMDV. Simulation results show that proposed algorithm consumes less energy towards given scenario. Simulations are performed in NS-2
A Study on Pollution Monitoring system in Wireless Sensor Networks
Get PDFAir pollution is one of environmental issues that cannot be ignored. The heavy transportation and urbanization result in the air pollutants concentrated in certain areas. Inhaling pollutants for a long time causes damages in human health. Traditional air quality monitoring methods, such as building air quality monitoring stations, are typically expensive. In addition, monitoring stations are generally less densely deployed and provide low resolution sensing data. This paper proposes an urban air quality monitoring system based on the technology of wireless sensor networks (WSNs). It also integrates with the global system for mobile communications (GSM). The system consists of sensor nodes, a gateway, and a control center managed by the Lab VIEW program through which sensing data can be stored in a database. This system is deployed to the main roads in the Taipei city to monitor the carbon monoxide (CO) concentration caused by vehicle emissions. The experimental results show that the proposed system is suitable for micro-scale air quality monitoring in real-time through the WSN technology
Data Mining in Biodata Analysis
Get PDFFor finding interesting patterns in large databases has lot of development in recent years.. Data mining is used in many fields like medicine, securing the data etc. Whereas bio data means the data regarding the biology, medical science, DNA technology and Bioinformatics in-depth analysis. Bio Informatics is the science which can perform managing, finding data, integrating, interrupting information from biological data, genomic, and metadata. Even additional knowledge and complexness can lead to the integration among genes. This paper is all about joining these two fields, the data regarding biology us ing data mining and gives the details of future developments in biodata analysis
On the Investigation of Biological Phenomena through Computational Intelligence
Get PDFThis paper is largely devoted for building a novel approach which is able to explain biological phenomena like splicing promoter gene identification disease and disorder identification and to acquire and exploit biological data This paper also presents an overview on the artificial neural network based computational intelligence technique to infer and analyze biological information from wide spectrum of complex problems Bioinformatics and computational intelligence are new research area which integrates many core subjects such as chemistry biology medical science mathematics computer and information science Since most of the problems in bioinformatics are inherently hard ill defined and possesses overlapping boundaries Neural networks have proved to be effective in solving those problems where conventional com-putation tools failed to provide solution Our experiments demonstrate the endeavor of biological phenomena as an effec-tive description for many intelligent applications Having a computational tool to predict genes and other meaningful in-formation is therefore of great value and can save a lot of expensive and time consuming experiments for biologists This paper will focus on issues related to design methodology comprising neural network to analyze biological information and investigate them for powerful application
Design and FPGA Implementation of High Speed DWT-IDWT Architecture with Pipelined SPIHT Architecture for Image Compression
Get PDFImage compression demands high speed architectures for transformation and encoding process Medical image compression demands lossless compression schemes and faster architectures A trade-off between speed and area decides the complexity of image compression algorithms In this work a high speed DWT architecture and pipelined SPIHT architecture is designed modeled and implemented on FPGA platform DWT computation is performed using matrix multiplication operation and is implemented on Virtex-5 FPGA that consumes less than 1 of the hardware resource The SPIHT algorithm that is performed using pipelined architecture and hence achieves higher throughput and latency The SPIHT algorithm operates at a frequency of 260 MHz and occupies area less than 15 of the resources The architecture designed is suitable for high speed image compression application
Performance Analysis of a Two Node Tandem Communication Network with Feedback
Get PDFA Communication Network needs optimal utilization of resources such as bandwidth routers transmitters etc In this paper we have developed and analyzed a communication network with two nodes with feedback In this network the arrival of packets characterized by homogeneous Poisson process and transmission of both the transmitters is characterized by Poisson process Dynamic bandwidth allocation policy is proposed by adjusting the transmission rate at every transmitter just before transmission of each packet The model is evaluated using the difference-differential equations and a probability generating function of the number of packets in the buffer Through mathematical modeling performance measures including average number of packets in each buffer the probability of emptiness of the network the average waiting time in the buffer and in the network the throughput of the transmitters utilization and the variance of the number of packets in the buffer are derived under transient condition
Security of Side Channel Power Analysis Attack in Cloud Computing
No full textFuture world is moving towards cloud computing In a word cloud computing can be defined as remote access computing resources through the Internet It provides a lot of services at a very low cost It can improve an organization s performance by minimizing its cost It also helps to maximize their revenue with fewer resources It is becoming popular among organizations and people They can store their data in cloud at low cost Attackers aim to attack cloud environment for getting valuable information from cloud users They attack on it by taking different approaches They try to access confidential information of different organizations from the cloud Among different attacks side channel power analysis attack is a newer type of attack In this paper we proposed a way to mitigate these types of attacks through a Police Virtual Machine Police VM The Police VM provides false power consumption information to attackers and they cannot get real power consumption information from user V
A Framework for Context-Aware Semi Supervised Learning
Get PDFSupervised learning techniques require large number of labeled examples to build a classifier which is often difficult and expensive to collect. Unsupervised learning techniques, even though do not require labeled examples often form clusters regardless of the intended purpose or context. The authors proposes a semi supervised learning framework that leverages the large number of unlabeled examples in addition to limited number of labeled examples to form clusters as per the context. This framework also supports the development of semi supervised classifier based on the proximity of unknown example to the clusters so formed. The authors proposes a new algorithm namely 201C;Semi Supervised Relevance Feature Estimation201D;, (SFRE), to identify the relevant features along with their significance weightages which is integrated with the proposed framework. Experiments conducted on the benchmark datasets from UCI gave results which are very promising and consistent even with lesser number of labeled examples