Print E-mail

Off-Line Arabic Handwritten Word Segmentation Using Rotational Invariant Segments Features

Shubair Abdulla1, Amer Al-Nassiri2, and Rosalina Abdul Salam3

1Faculty of Education and Basic Sciences, Ajman University of Science and Technology, UAE

2Faculty of IT, Ajman University of Science and Technology, UAE

3School of Computer Sciences, University SaMalaysia, Malaysia


Abstract: This paper describes a new segmentation algorithm for handwritten Arabic characters using Rotational Invariant Segments Features (RISF). The algorithm evaluates a large set of curved segments or strokes through the image of the input Arabic word or subword using a dynamic feature extraction technique then nominates a small “optimal” subset of cuts for segmentation. All the directions of stroke are converted to two main segments: '+' and w'-' RISF. A list of nominated segmentation points are prepared from the '+' segments and evaluated according to special conditions to locate the final segmentation points. The RISF algorithm was tested by using our new designed database AHD/AUST and the IFN/ENIT database. It has achieved a high segmentation rate of 95.66% on AHD/AUST and 90.58% on IFN/ENIT handwritten Arabic databases.

Keywords:  Feature extraction, Arabic character segmentation, cursive writing, Arabic words database.

Received June 24, 2006; accepted March 19, 2007

  Full Text                                                           
Print E-mail

A Heterogeneous Framework for the Global Parallelisation of Genetic Algorithms

Mohammad Hamdan

Department of Computer Science, Yarmouk University, Jordan

Abstract: There is a big need for the parallelisation of genetic algorithms.  In this paper, a heterogeneous framework for the global parallelisation of genetic algorithms is presented. The framework uses a static all-worker parallel programming paradigm based on collective communication. It follows the single program multiple data parallel programming model. It utilises the power of parallel machines by allowing multiple crossover and mutation operators being used within a single genetic algorithm.  This mixture of operators  can  be  applied  to  the  strings  of  a  population  in  parallel  without  changes  to the  canonical  sequential  genetic  algorithm. These features help the parallel genetic algorithm in exploiting the search space efficiently and thoroughly when compared to the sequential genetic algorithm. The framework is instantiated with specific parameters to solve an NP-hard problem, the asymmetric travelling salesman problem.  The results  for the  parallel  genetic  algorithm  are very  good  in  terms  of  solution  quality. Also very good speedup and scalability results were achieved on the parallel machine.

Keywords: Genetic algorithms, parallel processing, parallel genetic algorithms, crossover, mutation, TSP.

Received November 14, 2006; accepted February 28, 2007

Full Text
Print E-mail

Self Ranking and Evaluation Approach for 

Focused Crawler Based on Multi-Agent System

Hesham Ali

 Computers Engineering and Systems Department, Mansoura University, Egypt

Abstract: The need of better way of retrieving information and dealing with the increasing complexity and volume of information for users is an important research theme. Retrieving information from the www via search engine may be deliberate as the most significant one. Most of the recent efforts that had been done in this area suggest a better solution for general-purpose search engine limitations. That leads to a new generation of search engines called vertical-search engines. However, all current focused crawlers; (crucial component within vertical search engine) crawling strategies introduced high penalties as they consume network bandwidth and resources of hosting servers. This paper introduces a new approach for focused crawling that integrates evidence from both focused crawling and intelligent multi-agent technology. It will help in distributing the computation among data sites, hence overcoming the drawbacks of traditional approaches. We evaluate its performance compared to the client/server-based computation model from perspectives of amount of data transfer and execution time through both simulation and analytical study. The results achieved from the experiments confirm efficiency of the proposed approach. Both of execution time and traffic are reduced in the case of mobile agent compared with client/server model (i.e., eliminating the bottleneck, and overhead problem at crawler site). Furthermore, such combination will also simplify the design and implementation of focused crawlers.

Keywords: Search engine, vertical search engine, focused crawler, mobile agents, multi-agents systems.

Received December 12, 2006; accepted February 27, 2007

Full Text
Print E-mail

A Priority Gated Round Robin Polling

Scheme for Bluetooth Piconets

Asmaa Ibrahim, Rawya Rizk, and Gamal Mahmoud

Electrical Engineering Department, Suez Canal University, Egypt

Abstract: Bluetooth is a wireless access technology where polling is used to share bandwidth among the nodes. In this paper, a new polling scheme for intra-piconet scheduling in Bluetooth piconets, Priority Gated Round Robin, is proposed. The performance of this algorithm is analyzed via simulation in different cases such as different Segmentation And Reassembly algorithms, different packet sizes, and variable number of slaves and is compared with the traditional Gated Round Robin algorithm. Simulation results demonstrate that the new scheme achieves better performance over the existing Gated Round Robin scheme.

Keywords: Bluetooth, piconet, polling, Segmentation And Reassembly (SAR).

Received October 10, 2006; accepted February 25, 2007    


Print E-mail

Effect of Filter Size on Fusion Function in Information Retrieval

Nagammapudhur Gopalan 1 and Krishnan Batri 2

1Department of Computer Applications, National Institute of Technology, India

2Department of Computer Science and Engineering, National Institute of Technology, India

Abstract: An attempt to detach the worst performing schemes for improving the effectiveness of the information retrieval system, which combines the multiple sources of evidence, has been presented in this paper. The disturbances caused by the ill performing retrieval schemes are studied by using the concept of filter. The size of the filter is altered to examine the efficiency of the retrieval system. The experiments are conducted over the three-benchmark test collections viz., ADI, CISI and MED. The results indicate that, the efficiency of the retrieval system varies with the filter size and the maximum improvement in performance is achieved only at particular filter size, which is different for various fusion functions.

Keywords: Information retrieval, fusion, filter, overlap, and precision.

Received December 6, 2006; accepted February 21, 2007

Full Text
Print E-mail

Mobility Prediction Based Resource Reservation for Handoff in Multimedia Wireless Cellular Networks

Maher Al-Sanabani, Shamala Subramaniam, Mohamed Othman, and Zuriati Zukarnain

 Department of Communication Technology and Networks, University Putra Malaysia, Malaysia

Abstract: In this paper, a mobility predictive based resource (bandwidth) reservation is proposed to provide flexible usage of limited resource in wireless cellular networks. Currently, multimedia communications is urgently expected in wireless networks. That as there is more and more high speed applications that emerge for wireless networks. One of the most important and complicated issues is the Quality of Service (QoS) in next wireless networks. In other words, the problem to maintain the playing continuity of multimedia streams during handoff is hard for solution. Hence, in this paper, the scheme is proposed to provide high accurate prediction of next crossing cell that Mobile Terminal is going to, in order to avoid too early or over reservation resulting in a waste of resources. The amount of bandwidth to be reserved is dynamically adjusted according to the current position and extrapolated direction of Mobile Terminal. An admission control scheme is also considered to further guarantee the QoS of real time traffic. The performance of system is evaluated through discrete event simulation of realistic wireless cellular environment. Simulation results show that the proposed scheme as compared to the existing schemes is efficient to reduce the New Call Blocking Probability and the Handoff Call Dropping Probability of real time traffic. In addition, it is efficient to reduce the number of terminated ongoing calls of non-real time traffic and the cancels reservation that increase the system bandwidth utilization.

Keywords: Wireless cellular networks, mobility prediction, resource reservation, handoff and quality of service.

Received October 12, 2006; accepted March 25, 2007

Full Text
Print E-mail

An Efficient Region-Based Approach for Object Recognition and Retrieval Based on Mathematical Morphology and Correlation Coefficient

Jehad Alnihoud

Department of Computer Science, Al-al Bayt University, Jordan

Abstract: In this paper, a new technique based on slicing the images to equally sub-areas, then applying the density slicing to the color histogram of these areas combined with the color pair technique is presented as a  filtration phase based on the color features of images. This allows dramatic reduction in the candidate sets of images, which needs to be proceed through the time-consuming process of shape feature extraction and comparison algorithms. After that, shape recognition method is proposed as a higher level phase in the proposed system. The proposed approach for shape-based retrieval relies on detecting the edges of the objects, using laplacian of gaussian, in the image query as well as the set of filtered images from the proceeding layer of color approach. Next, the morphological techniques, using dilate and erosion are applied, in order to solidify the objects within the image and to increase the chance of correlation coefficient to be used as means of similarity between images. Furthermore, transformation of objects is applied to overcome the rotation based problem. The experimental results show that the proposed approach is fast, efficient, and feasible.

Keywords: Morphological operations, density slicing, color histogram, correlation coefficient.

Received September 27, 2006; accepted February 15, 2007

Full Text

Print E-mail

Simplified Neural Model for the Software Development Team Optimization

Madhu S. Nair and Jaya Vijayan

Rajagiri School of Computer Science, Rajagiri College of Social Sciences, India.

Abstract: A simplified neural model to optimize a project team and to attain maximum throughput as well as to obtain high quality software has been proposed here. A novel approach, which uses the concept of Artificial Neural Networks, to train the software professionals and make them perform at high level of standards, is adopted. In this approach, a high level of communication among the professionals is achieved which will lead to good team work and finally produce quality software that meets the required level of standard.

Keywords: Neural network, optimization, training, weights, acceptance.

Received August 8, 2006; accepted February 15, 2007

Full Text


Print E-mail

From e-Manufacturing to M-Manufacturing

Faiza Bouchoul and Mohamed Mostefai

 Laboratoire d’Automatique de Setif-LAS- Automation Lab. of Setif, Ferhat Abbas Uni., Algeria

Abstract: This paper studies the impact of wireless technologies on enterprises. A multitude of new paradigms and new software technologies have emerged from the web. E-business and e-manufacturing have really transformed concepts and practices in enterprises. Wireless technologies added new possibilities for these technologies, by this work we propose m-manufacturing as a component of e-manufacturing when practiced in mobile environment, then compare e-manufacturing to m-manufacturing. We applied our proposed approach to model m-maintenance of spare parts. Combination of mobile agent technology and fault-tolerant principles has given an appropriate global solution tom m-maintenance issues

Keywords: Wireless technologies, mobile enterprise, m-manufacturing, m-maintenance, mobile agents, fault-tolerance

Received May 31, 2006; accepted December 27, 2006

Full Text


Print E-mail

Medical Image Compression Using a Novel Embedded Set Partitioning Significant and Zero Block Coding

Geetha Palanisamy1 and Annadurai Samukutti2

1Department of CSE & IT, Anna University, India

2Government College of Engineering, India

Abstract: Lossless compression with progressive transmission is playing a key role in telemedicine applications. The proposed lossless embedded progressive region based significant and zero block coding uses a recursive set partitioning procedure to sort subsets of wavelet coefficients by maximum magnitude with respect to thresholds that are integer powers of two. It exploits two fundamental characteristics of a wavelet transform – the well defined hierarchical structure and energy clustering in transform domain. The proposed low complexity coder is rooted from Set Partitioning Embedded Block (SPECK) and it is a distortion scalable coder which achieves excellent coding performance. Extensive experiments have verified that the proposed coder is competitive with other state of the coders such as set partitioning in hierarchical trees, SPECK, JPEG2000 on compression performance and computational complexity for standard test images.

Keywords: ESPSZBC, lossless compression, progressive transmission, SPECK, SPIHT.

Received November 4, 2006; accepted November 24, 2006

Full Text

Print E-mail

Performance Evaluation of Gene Expression Programming for Hydraulic Data Mining

Khalid Eldrandaly1 and Abdel-Azim Negm2

1Information Systems Department, College of Computers, Zagazig University, Egypt

2Water Engineering Department, College of Engineering, Zagazig University, Egypt

Abstract: Predication is one of the fundamental tasks of data mining. In recent years, Artificial Intelligence techniques are widely being used in data mining applications where conventional statistical methods were used such as Regression and classification. The aim of this work is to show the applicability of Gene Expression Programming (GEP), a recently developed AI technique, for hydraulic data prediction and to evaluate its performance by comparing it with Multiple Linear Regression (MLR). Both GEP and MLR were used to model the hydraulic jump over a  roughened bed using very large series of experimental data that contain all the important flow and roughness parameters such as the initial Froude number, the height of roughness ratio, the length of roughness ratio, the initial length ratio (from the gate) and the roughness density. The results show that GEP is a promising AI approach for hydraulic data prediction.

Keywords: Data mining, multiple linear regression, MLR, gene expression programming, GEP, hydraulic jump.

Received August 24, 2006; accepted November 23, 2006

Print E-mail

Modularization of Crosscutting Concerns in Requirements Engineering

Abdelkrim Amirat1, 2, Mohamed Laskri1, and Tahar Khammaci2

1 Laboratoire de Recherche en Informatique, Université de Annaba, Algeria

2Laboratoire LINA CNRS FRE 2729, Université de Nantes 2, France

Abstract: In spite of the generated benefits, Object-Oriented (OO) paradigm seems reaching its limits, regarding complexity reduction of current systems. In this context, the Aspect Oriented (AO) comes up as an alternative to reduce software development complexity while keeping OO advantages. Needs for investigating methodologies of AO Software Development have emerged a long with AO. As an example, Early Aspect (EA) aims to identify aspects on the early stages of software development, such as domain analysis requirements specification and architectural design. Being one of the newest software engineering paradigms, AO emphasizes that new studies and experiments should be carefully carried out, in order to establish improved methods, techniques and tools applicable to this new way of development. In this paper, we discuss a sequence of systematic activities toward an early consideration of specifying and separating crosscutting Functional Requirements (FRs) and Non-Functional Requirements (NFRs) by the adoption of use-cases to model systems. This approach would make it possible to identify and resolve conflicts between requirements earlier in the development cycle and can promote traceability of broadly scoped properties throughout system development, maintenance and evolution.

Keywords: Early aspects, requirements engineering, decomposition, composition, methodologies.

Received May 30, 2006; accepted November 22, 2006

Full Text

Print E-mail

Performance Analyses of Fast Frequency Hopping Spread Spectrum and Jamming Systems

Abid Yahya1, Othman Sidek1, and Junita Mohamad-Saleh2

1Collaborative MicroElectronic Design Excellence Center, Universiti Sains Malaysia, Malaysia

2School of Electrical and Electronic Engineering, Universiti Sains Malaysia, Malaysia

Abstract: As technology becomes increasingly able to meet the requirements, interest in faster, noncoherent, frequency hopping rates to reduce the jamming of communication has heightened. The focus of this paper is on the performance of the fast frequency hopping spread spectrum system operating in the presence of partial band noise jamming. In this paper we consider a communication system that transmits binary data sequence or M frequency shift keying over a channel. With non-coherent detection, the MFSK tones on a given hop must be separated in frequency by an integer multiple of chip rate to provide orthogonality. The worst case partial-band noise jammer chooses fraction () to maximize the bit error probability (Pb) for a given M and signal to noise ratio. It has been observed from the simulation results that increases with K, illustrating the effectiveness of worst-case jamming () against FH/MFSK signals at typical operating points. It may be noted that   decreases as Eb/NJ gets larger. <!--[if !vml]--><!--[endif]--> <!--[if !vml]--><!--[endif]--> <!--[if !vml]--><!--[endif]-->

Keywords: Frequency hopping spread spectrum, interference, partial band noise jamming.

Received August 23, 2006; accepted February 19, 2007

Print E-mail

Strategic Planning for e-Development in Developing Countries

Abdel Nasser Zaied

College of Graduate Studies, Arabian Gulf University, Bahrain

Abstract: One of the most persistent criticisms of open markets and free trade is that they increase the gap between developed and developing countries. The rapid emergence of the internet economy is giving this argument new impetus and is increasing focus on the digital divide and e-development to reduce the gap between “information rich” and “information poor” countries. Information and Communication Technology (ICT) is widely recognized as a potentially powerful tool to minimize this gap. ICT can provide a great opportunity for developing countries to leapfrog stages of economic development. A strategic management approach to ICT is therefore necessary in providing national guidelines on how ICT can be harnessed by developing countries to increase their technological capabilities towards e-development.This  study  proposes  a framework  that  aimed  at   facilitating  the  process  of  creating  and  implementing   ICT   strategies. This framework which comprises three phases is a step-by-step methodology that can help the developing countries to establish their own ICT strategies. It needs further studies to complement with countries mission based on their resources and situations.

Keywords: Strategic planning, e-development.

Received July 1, 2006; accepted November 21, 2006

Full Text
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/ on line 251 Warning: fsockopen(): unable to connect to (php_network_getaddresses: getaddrinfo failed: Name or service not known) in /hsphere/local/home/ccis2k/ on line 251 skterr