May 2015. No.3
Print E-mail

Face Recognition using Truncated Transform Domain Feature Extraction

Rangan Kodandaram 1, Shashank Mallikarjun 1 Manikantan Krishnamuthan 1and, Ramachandran Sivan2

1Ramaiah Institute of Technology, Bangalore, India

2S.J.B Institute of Technology, Bangalore, India

 Abstract: Face Recognition (FR) under varying pose is challenging and exacting pose invariant features is an effective approach to solve this problem. In this paper, we propose a novel Truncated Transform Domain Feature Extractor (TTDFE) to improve the performance of the FR system. TTDFE involves a unique combination of   Symlet - 4 DWT, 2D - DCT, followed by a novel truncation process. The truncation process extracts higher amplitude coefficients from the Discrete Cosine Transform (DCT) matrix. An optimal truncation point is estimated, which is inspired by a relationship developed between the image dimensions and the positions of DCT amplitude peaks. TTDFE is used for efficient feature extraction and a Binary Particle Swarm Optimization - based feature selection algorithm is used to search the feature space for the optimal feature subset. Experimental results, obtained by applying the proposed algorithm on 5 benchmark  face databases with large pose variations, namely FERET, UMIST, FEI, PHPD and IFD, show that the proposed system outperforms other FR systems. A significant increase in the Recognition Rate and a substantial reduction in the number of features selected, are observed.

 Keywords: FR, feature extraction, discrete wavelet transform, discrete cosine transform, feature selection, binary particle swarm optimization.

 Received January 28, 2013; accept March 2, 2014

Full Text

 

 

 
Print E-mail

An Efficient ROI Encoding Based on LSK and Fractal Image Compression

TMP Rajkumar1 and Mrityunjaya V Latte2

1Research Scholar, Anjuman Engineering College, India

2Principal, JSS Academy of Technical Education, India

 

Abstract: Telemedicine is one of the emerging fields in medicine which is characterized by transmitting medical data and images between different users. The medical images which are transmitted over the internet require huge bandwidth. Even images of single patient are found to be very huge in size due to resolution factor and number of images per diagnosis. So there is an immense need for efficient compression techniques that can be used to compress these medical images. In medical images, only some of the regions are considered to be more important than the others (e.g., Tumor in brain Magnetic Resonance Imaging (MRI)). This paper reviews the application of ROI coding in the field of telemedicine. The image coding is done using wavelet transform based on Listless SPECK (LSK). The ROI is obtained from user interaction and coded with the user given resolution to get high compression ratio. In our proposed method, instead of decompressing all the blocks, we decompress only the similar blocks based on the index valued stored on the stack. Thus our proposed method efficiently compresses the medical image. The performance measure can be analyzed by using PSNR. The execution time of the proposed method will be reduced when compare to the other existing methods. The experimental result shows that the application of ROI coding using LSK brings about high compression rate and quality ROI.

 

Keywords: Image compression, ROI, LSK, fractal image compression, MRI images, Iterated Functions Systems (IFS).

 

Received January 21, 2012; accept September 30, 2013

Full Text

 

 

 
Print E-mail

An Efficient Content Based Image Retrieval Using Advanced Filter Approaches

Manoharan Subramanian1 and Sathappan Sathappan2

1Department of Computer Science (PG), Kongu Arts and Science College, India

 2Department of Computer Science, Erode Arts and Science College, India

Abstract: In general the users are in need to retrieve images from a collection of database images from variety of domains. In earlier phase this need was satisfied by retrieving the relevant images from different database simply. Where there is a bottleneck that the images retrieved was not relevant much to the user query because the images were not retrieved based on content where another drawback is that the manual searching time is increased. To avoid this Content - Based Image Retrieval (CBIR) is developed it is a technique for retrieving images on the basis of automatically - derived features such as colour, texture and shape of images. To provide a best result in proposed work we are implementing high level filtering where we are using the Anisotropic Morphological Filters, hierarchical  Kaman  filter and particle filter proceeding with feature extraction method based on color and gray level  feature  and after this the results were normalized.

 

Keywords: CBIR, anisotropic morphological filters, hierarchical kaman filter and particle filter, mahalanobis distance, color feature extraction, gray-level extraction.

 Received January 13, 2013; accept  January 29, 2014

Full Text

 

 

 
Print E-mail

Arabic Phonemes Transcription Using Data Driven Approach

Khalid Nahar1, Husni Al-Muhtaseb1, Wasfi Al-Khatib1, Moustafa Elshafei1 and Mansour Alghamdi2

1College of Computers and Information Technology, Tabuk University, Saudi Arabia

2Computer Research Institute, King Abdulaziz City for Science and Technology (KACST),

Saudi Arabia

Abstract: The efficiency and correctness of continuous Arabic Speech Recognition Systems (ARS) hinge on the accuracy of the language phoneme set. The main goal of this research is to recognize and transcribe Arabic phonemes using a data-driven approach. We used the Hidden Markov Toolkit (HTK) to develop a phoneme recognizer, carrying out several experiments with different parameters, such as varying number of Hidden Markov Model (HMM) states and Gaussian mixtures to model the Arabic phonemes and find the best configuration. We used a corpus consisting of about 4000 files, representing 5 recorded hours of modern standard Arabic of TV - News. A statistical analysis for the phonemes length, frequency and mode was carried out, in order to determine the best number of states necessary to represent each phoneme. Phoneme recognition accuracy of 56.79% was reached without using a language model. The recognition accuracy increased to 96.3% upon using a bigram language model.

 

Keywords: Phoneme transcription, data - driven, speech recognition, network lattices, Arabic speech corpus.

Received November 20, 2012; accept March 21, 2014

Full Text

 

 

 
Print E-mail

Efficient English Auction Scheme without a Secure Channel

Tzong-Chen Wu1, Tzuoh-Yi Lin1, Tzong-Sun Wu2,*, and Han-Yu Lin2

1Department of Information Management, National Taiwan University of Science and Technology, Taiwan

2Department of Computer Science and Engineering, National Taiwan Ocean University, Taiwan

Abstract: English auctions become tremendously popular on the Internet today. This paper presents a new English auction scheme that can be realized in the public network environments without any additional secure channel. Our scheme not only satisfies security requirements of anonymity, traceability, no framing, fairness, public verifiability, unlinkability among different rounds of auction, and linkability in an auction round, but also provides one-time registration and easy revocation. Furthermore, as compared with the pervious works, the proposed scheme has a better performance in terms of the computation and the size of bidding information.

 

Keywords: English auction, bilinear pairings, bilinear diffie-hellman problem, hash function.

 

Received August 6, 2012; accepted March 23, 2014

Full Text

 

 

 

 
Print E-mail

Anomaly Traffic Detection Based onPCA and SFAM

Preecha Somwang1, 2 and Woraphon Lilakiatsakun2

1Office of Academic Resources and Information Technology, Rajamangala University of Technology Isan, Thailand

 2Faculty of Information Science and Technology, Mahanakorn University of Technology, Thailand

Abstract: Intrusion Detection System (IDS) has been an important tool for network security. However, existing IDSs that have been proposed do not perform well for anomaly traffics especially Remote to Local (R2L) attack which is one of the most concerns. We thus propose a new efficient technique to improve IDS performance focusing mainly on R2L attacks. The Principal Component Analysis (PCA) and Simplified Fuzzy Adaptive Resonance Theory Map (SFAM) are used to work collaboratively to perform feature selection. The results of our experiment based on KDD Cup’99 dataset show that this hybrid method improves classification performance of R2L attack significantly comparing to other techniques while classification of  the other types of attacks are still well performing.

 Keywords: Intrusion detection system; network security; PCA; SFAM.

 Received May, 3, 2013; accepted March, 24, 2014

Full Text

 

 

 

 

 
Print E-mail

Multi Dimensional Taxonomy of Bio-inspired Systems Based on Model Driven Architecture

Seif Mili and Djamel Meslati

LISCO Laboratory, Badji Mokhtar-Annaba University, Algeria

 Abstract: The biological metaphor is an analogy between the biological world and the artificial world that enables us to benefit  from artificial approaches by imitating some biological aspects while ignoring others. The biological metaphors, also called bio-inspired approaches, depend not only, on the biological  field considered, but also on our understanding of that field and the paradigms and means we use to extract practical and useful elements to model some aspects of that field. Today, there is a huge number of metaphors which are different by their very nature and this number is expected to increase according to our inspiration capabilities. In front of this increasing numbers of metaphors it becomes necessary to define the main features of each one in order to evaluate their practical impact, to compare them, to ease their learning and use, to combine them, etc. Finding the main or common features of bio-inspired approaches is not an easy task. Although significant achievement can be noticed in some fields like artificial neural networks or genetic algorithms, a common agreement on definitions and concepts of a huge number of bio-inspired approaches is still lacking. In this article, we propose a multi-dimensional approach based on the Model Driven Architecture (MDA) to describe conceptually a wide range of bio-inspired approaches. Our starting point is to consider that each bio-inspired approach has two aspects: Structural aspect and Behavioural aspect. While the structural aspect is concerned with the involved elements and their relationships, the behavioural aspect deals with the process by which a computing is achieved in an artificial system based on the considered bio-inspired approach. Our choice of the MDA paradigm is justified by its ability to describe uniformly various intricate processes and artefacts involved in the development of software systems. As a preliminary result, our description approach proved to be effective in characterizing a wide range of bio-inspired systems.

 Keywords : Bio-inspired system taxonomy, MDA, ontogeny, phylogeny, epigeny.

 Received September 20, 2012; accepted October3, 2013

 

 

 
Print E-mail

Novel Approaches for Scheduling Task Graphs in Heterogeneous Distributed Computing Environment

 

Ehsan Munir1, Saima Ijaz1, Sheraz Anjum1, Ali Khan3, Waqas Anwar2 and Wasif Nisar1

1Department of Computer Science, COMSATS Institute of Information Technology, Wah Cantt, Pakistan

2Department of Computer Science, COMSATS Institute of Information Technology, Abbottabad,   Pakistan

3Department of Electrical Engineering, COMSATS Institute of Information Technology, Lahore, Pakistan

 

Abstract: Distributed heterogeneous computing environment comprises of diverse set of interconnected resources that are capable of performing computationally complex tasks efficiently. In order to exploit the high performance of such a system, the task scheduling problem demands for the efficient mapping of the tasks. Because of its fundamental importance, the problem has been studied extensively and several algorithms have been proposed. In this paper, we propose two novel approaches for the task scheduling problem and compare the proposed work on the basis of randomly generated task graphs with the well-known existing algorithms. The simulation results elucidate on the basis of different cost and performance metrics that for most of the scenarios, the proposed approaches outperform the existing ones considerably.


Keywords: Heterogeneous distributed computing systems (HDCS), directed acyclic graphs (DAG), task scheduling, task prioritization, makespan.

 Received April 12, 2012; accept April 1, 2014

 Full Text

   

 

 
Print E-mail

A Hierarchical Approach to Improve Job Scheduling and Data Replication in Data Grid

Somayeh Abdi1 and Sayyed Mohsen Hashemi2

*,1Faculty of Computer Engineering, Islamic Azad University, Iran

2Faculty of Computer Engineering, Islamic Azad University, Iran

 Abstract: In dynamic environment of Data Grid effective job scheduling methods consider location of required data in dispatching jobs to resources. Also job scheduling methods are combined with data replication mechanisms to reduce remote data access as well as save network bandwidth. In this paper we combine job scheduling method and dynamic data replication to reduce data access delay and job execution time. Also we expand our work by applying Bloom Filter in job scheduling decision. In Data Grid, appropriate mechanisms for recording, deleting and inquiring information about data files are required for implementing proper job scheduling method. Therefore we apply counting Bloom Filter for recording/deleting and inquiring information about data files in replica catalogue. Result of simulation indicates that proposed job scheduling and data replication methods reduce job execution time, also using Bloom Filter saves network bandwidth and reduces time of gathering information for selecting appropriate resources in job scheduling.

Keywords: Job scheduling, data replication, replica catalogue, replica manager, resource discovery, counting bloom filter.

 

Received August 8, 2012; accept April 24, 2014

Full Text

 

 

 

 
Print E-mail

A Comparative Analysis of Software Protection Schemes

Muhammad Khan1, Muhammad Akram2, and Naveed Riaz3

1,2Shaheed Zulfikar Ali Bhutto Institute of Sciences and Technology, Islamabad, Pakistan

3College of Computer Science and Information Technology, University of Dammam, Saudi Arabia

 

Abstract: In the era of software globalization, the need for securing software is much sought to ensure its smooth functioning for continuous availability of services to the users. Particularly, in cloud computing environment, all the software in the cluster needs to be secured and shielded against unauthorized accesses. Software crackers are always in the search of flaws in the software to obtain access to the software functionally by penetrating into the software skeleton. This paper reviews and critically analyzes various software protection techniques, both software-based and hardware-based, that can help control the software piracy issues in order to determine their efficacy and specific use in different environments and scenarios. The software protection techniques explored in this paper include cryptography, software watermarking, secure access scheme, software aging, guards, obfuscation and multi-block hashing techniques. The paper also discusses the taxonomy of the software protection techniques and the probable attack models that can be launched against each technique to evade the protection mechanism.

Keywords: Software tempering, software reversing, cryptography, watermarking, digital rights management.

Received March 21, 2013; accept April 2, 2014

Full Text

 

 

 

 

 
Print E-mail

 

Stability Coalition Formation with Cost Sharing in

Multi-Agent Systems Based on Volume Discount

Walaa El-Ashmawi1, Hu Jun1, 2 and Li Renfa1

1College of Information Science and Engineering, Hunan University, China

2State Key Laboratory for Novel Software Technology, Nanjing University, China

 Abstract: In Multi-Agent Systems(MAS), cooperation among agents to form coalitions based on volume discounts is a key topic. Such cooperation enables agents to achieve goals that they may not have been able to achieve independently at lower prices without ordering more than their real demand. In this paper, we propose a Stability Coalition Formation (SCF) and payoff distribution in terms of the core. Agents can enjoy a price discount for each of their requested action to achieve a goal through the concept of Social Agent Networ (SAN)k, where different opportunities can be found. Each opportunity is associated with coalition value and search cost, given a search cost, the goal of the agent is to find the best set of opportunities which fulfills the coalition’s demands, along with a cost sharing rule satisfying certain stability properties. The experimental results illustrated that, the performance of proposed semi-optimal solution to SCF has proven its stability with average payoff 99.98% closest to the optimal payoff and higher than the average coalition value obtained by 9% when considered a search cost as a parameter affected on the search for optimal coalitions. Also, it has proven its efficiency in average processing time that saved and reduced by 15%~44% according to a different number of agents.

 Keywords: MAS, coalition formation, volume discount, search cost, stability.

Received March 20, 2013; accept April 5, 2014

 Full Text

 
Print E-mail

A Human Activity Recognition System Using Hidden Markov Models with Generalized Discriminant Analysis on Enhanced Independent Component Features

1Md. Zia Uddin, 2*Deok-Hwan Kim, 3Tae-Seong Kim

1 Department of Computer Education, Sungkyunkwan University, Republic of Korea.

2School of Electronic Engineering, Inha University, Republic of Korea

3Department of Biomedical Engineering, Kyung Hee University, Republic of Korea

Abstract:  Human activity recognition from time-sequential video images is an active research area in various applications such as video surveillance and smart homes nowadays. This paper presents a novel approach of automatic human activity recognition based on Generalized Discriminant Analysis (GDA) on Enhanced Independent Component (EIC) features from binary silhouette information to be used with Hidden Markov Model (HMM) for training and recognition. The recognition performance using GDA on EIC features has been compared to other conventional approaches including Principle Component (PC), EIC, and Linear Discriminant Analysis (LDA) on PC features where the preliminary results show the superiority of the proposed approach.

Keywords:  Human activity recognition, EICA, GDA, and HMM.

Received May 13, 2013; accept April 7, 2014

Full Text

 

 
Print E-mail

SemanticBoolhean Arabic Information Retrieval

 

Emad Elabd, Eissa Alshari and Hatem Abdulkader

Faculty of computers and information, Menoufia University, Egypt

 

Abstract: Arabic language is one of the most widely spoken languages. This language has a complex morphological structure and is considered as one of the most prolific languages in terms of article linguistic. Therefore, Arabic Information Retrieval (AIR) models need specific techniques to deal with this complex morphological structure. This paper aims to develop an integrate AIR frameworks. It lists and analysis the different Information Retrieval (IR) methods and techniques such as query processing, stemming, and indexing which are used in AIR systems. We conclude that Arabic information retrieval frameworks have a weakness to deal with semantic in term of Indexing, Boolean model , Latent Semantic Analysis (LSA), Latent semantic Index (LSI), and semantic ranking. Therefore, semantic Boolean IR framework is proposed in this paper. This model is implemented and the precision, recall and run time are measured and compared with the traditional IR model.

 Keywords: AIR, semantic web, arabic language, ontology.

 

 Received August 15, 2013; accept April 13, 2014

 
<
 
Copyright 2006-2009 Zarqa Private University. All rights reserved.
Print ISSN: 1683-3198.
 
 
Warning: fsockopen(): php_network_getaddresses: getaddrinfo failed: Name or service not known in /hsphere/local/home/ccis2k/ccis2k.org/iajit/templates/rt_chromatophore/index.php on line 251 Warning: fsockopen(): unable to connect to oucha.net:80 (php_network_getaddresses: getaddrinfo failed: Name or service not known) in /hsphere/local/home/ccis2k/ccis2k.org/iajit/templates/rt_chromatophore/index.php on line 251 skterr