1Kalli Srinivasa Nageswara Prasad and 2Seelam Ramakrishna
1Research Scholar in Computer Science, Sri Venkateswara University, India
2Department of Computer Science, Sri Venkateswara University, India

Abstract: The rapid proliferation of Global Position Service (GPS) devices and mounting number of traffic monitoring systems employed by municipalities have opened the door for advanced traffic control and personalized route planning. Most state of the art traffic management and information systems focus on data analysis, and very little has been done in the sense of prediction. In this article, we devise an efficient system for the prediction of peak traffic flow using machine learning techniques. In the proposed system, the traffic flow of a locality is predicted with the aid of the geospatial data obtained from aerial images. The proposed system comprises of two significant phases: 1). Geospatial data extraction from aerial images and 2). Traffic flow prediction using See5.0 decision tree. Firstly, geographic information essential for traffic flow prediction are extracted from aerial images like traffic maps, using suitable image processing techniques. Subsequently, for a user query, the trained See5.0 decision tree predicts the traffic state of the intended location with relevance to the date and time specified. The experimental results portray the effectiveness of the proposed system in predicting traffic flow.

Keywords: Traffic flow, traffic prediction, spatial data mining, spatial data base, see5 decision tree algorithm.

 

Received January 9, 2012; accepted May 22, 2012

  

Full Text

Saed Alqaraleh and Omar Ramadan
Department of Computer Engineering, Eastern Mediterranean University, North Cyprus

Abstract: In this paper, we have proposed a new method for eliminating repeated occurrences in multimedia search engines. We have built software that extracts information from multimedia databases which will compare these multimedia files and marks only one copy of repeated files. The developed software can work with any search engine and can also work in a routine manner to deal with any updates on the databases. Moreover, the software allows multiple copies to be executed in parallel and consequently it improves the efficiency of multimedia searching.

Keywords: Search engines, multimedia search engines, information retrieval.

 

Received September 11, 2011; accepted December 30, 2011

  

Full Text

Baraa Sharef1,3, Nazlia Omar1, and Zeyad Sharef2
1Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia, Malaysia
2College of Electronic Engineering, University of Mosul, Iraq

3College of Computer Science and Mathematics, University of Mosul, Iraq

Abstract: Compared to other languages, there is still a limited body of research which has been conducted for the automated Arabic text categorization (TC) due to the complex and rich nature of the Arabic language. Most of such research includes supervised machine learning approaches such as Naïve Bayes, K-Nearest Neighbor, Support Vector Machine and Decision Tree. Most of these techniques have complex mathematical models and do not usually lead to accurate results for Arabic TC. Moreover, all the previous research tended to deal with the feature selection and the classification respectively as independent problems in automatic TC which led to the cost and complex computational issues. Based on this, the need to apply new techniques suitable for Arabic language and its complex morphology arises. A new approach in the Arabic TC term called the Frequency Ratio Accumulation Method (FRAM) which has a simple mathematical model is applied in this study. The categorization task is combined with a feature processing task. The current research mainly aims at solving the problem of automatic Arabic TC by investigating the Frequency Ratio Accumulation Method in order to enhance the performance of Arabic TC model. The performance of FRAM classifier is compared with three classifiers based on Bayesian theorem which are called Simple Naïve Bayes, Multi-variant Bernoulli Naïve Bayes and Multinomial Naïve Bayes models. Based on the findings of the study, the FRAM has outperformed the state-of-the-arts. It’s achieved 95.1% macro-F1 value by using unigram word-level representation method.

Keywords: Arabic text categorization, FRAM, automatic text categorization, text classification.

 

Received July 14, 2012; accepted December 6, 2012

  

Full Text

Nandhini Sivasubramaniam1 and Palaniammal Senniappan2
1Department of Computer Science, Garden City College of Science and Management, India
2Department of Science and Humanities, VLB Janakiammal College of Engineering and Technology, India

Abstract: The Core Stateless Fair Queuing (CSFQ) is distributed approach of fair queuing. The limitations include its inability to estimate fairness during large traffic flows which are short and bursty (VoIP or video) and also it utilizes the single FIFO queue at the core router. For improving the fairness and efficiency, we propose an Enhanced Core Stateless Fair Queuing (ECSFQ) with multiple queue priority scheduler. Initially priority scheduler is applied to the flows entering the ingress edge router. If it is real time flow i.e. VoIP or video flow, then the packets are given higher priority else lower priority.  In core router, for higher priority flows the multiple queue fair queuing is applied that allows a flow to utilize multiple queues to transmit the packets. In case of lower priority, the normal max-min fairness criterion of CSFQ is applied to perform probabilistic packet dropping. By simulation results, we show that this technique improves the throughput of real time flows by reducing the packet loss and delay.

Keywords: CSFQ, quality of service, multiple queue fair queuing, priority scheduler.

 

Received November 21, 2011; accepted December 18, 2012

  

Full Text

Saeed Amirgholipour Kasmani and Aboosaleh Mohammad Sharifi
Department of Computer, Ramsar Branch, Islamic Azad University, Ramsar, Iran

Abstract: In image processing Pre-processing is used for preparing or improving performance of operations. In order to improve performance of extraction algorithms in DCT based watermarking method, a new pre-filtering method is proposed in this paper. Enhancement filters are applied to the watermarked image as Pre-filtering before running watermark extraction algorithms in DCT based method. These filters are based of mixture of two filters: Unsharp and Laplacian of Gaussian (LoG). Distinction of watermarked part and unwatermarked part is increased by these filters; thus, the watermark information could be extracted with more accuracy. To show the effectiveness of the proposed method, different types of attacks are applied on typical DCT based algorithms. Experimental results show that extracted watermark has better quality than previous method.

Keywords: Digital image watermarking, watermark detection, DCT based watermarking, unsharp filter,  LoG filter.

 

Received February 5, 2012; accepted November 15, 2012

  

Full Text

Sanjay Singh1 and Trimbak Ramchandra Sontakke2
1Department of Computer Engineering, Siddhant College of Engineering, India
2Principal, Siddhant College of Engineering, India

Abstract: Nowadays, Content Based Image Retrieval (CBIR) plays a significant role in the image processing field. Images relevant to a given query image are retrieved by the CBIR system utilizing either low level features (such as shape, color etc.,) or high level features (human perception). Normally, a semantic gap exists between the low level features and the high level features, because the images which are identical in visual content may not be identical in the semantic sense. In this paper, an effective approach is proposed to trim down this semantic gap that exists between the low level features and the high level features. Initially, when a query image is given, images relevant to it are retrieved from the image database based on its low level features. We have performed retrieval utilizing one of the evolutionary algorithms called Evolutionary Programming (EP). Subsequent to this process, query keyword which is a high level feature is extracted from these retrieved images and then based on this query keyword, relevant images are retrieved from the database. Subsequently, the images retrieved based on low level features and high level features are compared and the images which are both visually and semantically identical are identified. Better results obtained by the proposed approach when it is queried using different types of images prove its effectiveness in minimizing the semantic gap.

Keywords: CBIR, low level feature, high level feature, semantic gap, color, shape, texture, contourlet, Squared Euclidean Distance (SED), EP.

 

Received February 5, 2012; accepted November 15, 2012

  

Full Text

Majid Masoumi1 and Shervin Amiri2
1Department of Electrical Engineering, Islamic Azad University Qazvin Branch, Iran
2Scientific Member of electrical engineering Department, Iranian Research Organization for Science and Technology, Iran

 

Abstract: Embedding a digital watermark into an electronic document is proving to be a feasible solution for multimedia copyright protection and authentication purposes. In the present paper we propose a new digital video watermarking scheme based on scene change analysis. By detecting the motion scene of video and using CDMA techniques the watermark is embedded into mid-frequency sub-bands of wavelet coefficients. In this experiment in order to enhance the security of our algorithm four keys are considered. Three of them are formed in watermark encryption process and one key is related to CDMA embedding process. Also, with the aim of making a good compatibility between the proposed scheme and Human Visual System (HVS), the blue channel of RGB video is utilized to embed the watermark.  Experimental results show the high robustness of the proposed method against both intentional and unintentional attacks during the transfer of video data. The implemented attacks are Gaussian noise, median filtering, frame averaging, frame dropping, geometric attacks and different kinds of lossy compressions including MPEG-2, MPEG-4, MJPEG and H.264/AVC.

Keywords: Digital watermarking, scene change analysis, geometric attacks, information security, HVS.

 

Received May 13, 2012; accepted December 30, 2012

Ilango Murugappan1 and Mohan Vasudev2
1Department of Computer Applications, K L N College of Engineering, India
2Department of Mathematics, Thiagarajar College of Engineering, India

 

Abstract: Data deals with the specific problem of partitioning a group of objects into a fixed number of subsets, so that the similarity of the objects in each subset is increased and the similarity across subsets is reduced. Several algorithms have been proposed in the literature for clustering, where k-means clustering and Fuzzy C-Means (FCM) clustering are the two popular algorithms for partitioning the numerical data into groups. But, due to the drawbacks of both categories of algorithms, recent researches have paid more attention on modifying the clustering algorithms. In this paper, we have made an extensive analysis on modifying the FCM clustering algorithm to overcome the difficulties possessed by the K-means and FCM algorithms over high dimensional data. According to, we have proposed an algorithm, called Projected Clustering based on FCM Algorithm (PCFA). Here, we have utilized the standard FCM clustering algorithm for sub-clustering high dimensional data into reference centroids. The matrix containing the reference values is then fed as an input to the modified FCM algorithm. Finally, experimentation is carried out on the very large dimensional datasets obtained from the benchmarks data repositories and the performance of the PCFA algorithm is evaluated with the help of clustering accuracy, memory usage and the computation time. The evaluation results showed that, the PCFA algorithm shows approximately 20% improvement in the execution time and 50% improvement in memory usage over the PCKA algorithm.

Keywords: Clustering, FCM, modified FCM, K-mean clustering, accuracy, memory usage, computation time.

 

Received November 29, 2011; accepted December 9, 2012

Faisal Ahmed1, Hossain Bari2, and Emam Hossain3
1Department of CSE, Islamic University of Technology, Bangladesh
2Samsung Bangladesh R & D Center Ltd, Bangladesh
3Department of CSE, Ahsanullah University of Science and Technology, Bangladesh

 

Abstract: Automatic recognition of facial expression is an active research topic in computer vision due to its importance in both human-computer and social interaction. One of the critical issues for a successful facial expression recognition system is to design a robust facial feature descriptor. Among the different existing methods, the Local Binary Pattern (LBP) has been proved to be a simple and effective one for facial expression representation. However, the LBP method thresholds P neighbors exactly at the value of the center pixel in a local neighborhood and encodes only the signs of the differences between the gray values. Thus, it loses some important texture information. In this paper, we present a robust facial feature descriptor constructed with the Compound Local Binary Pattern (CLBP) for person-independent facial expression recognition, which overcomes the limitations of LBP. The proposed CLBP operator combines extra P bits with the original LBP code in order to construct a robust feature descriptor that exploits both the sign and the magnitude information of the differences between the center and the neighbor gray values. The recognition performance of the proposed method is evaluated using the Cohn-Kanade (CK) and the Japanese Female Facial Expression (JAFFE) database with a support vector machine (SVM) classifier. Experimental results with prototypic expressions show the superiority of the CLBP feature descriptor against some well-known appearance-based feature representation methods.

Keywords: Facial expression recognition, feature descriptor, LBP, support vector machine, texture encoding.

 

Received April 26, 2012; accepted December 31, 2012

Dongfu Xie
College of Mechanical and Electrical Engineering, Jiaxing University, China

 

Abstract: The purpose of this paper is to present a soft decoding algorithm orienting to hardware implementaion for the (24, 12, 8) Golay code, and implement such an algorithm in field programming gates array (FPGA). The soft decoding algorithm devised by Lin’s et al. for the (24, 12, 8) is not suitable for hardware implementation because of involving many arithmetic operations, such as multiplications and divisions. To remove the complexity arithmetic operations, the absolute value of the channel information instead of the bit-error probability is employed to indicate the channel confidence. Moreover, the architecture developed for realizing the proposed algorithm is verified in a FPGA prototype. The BER performance obtained by the proposed decoding algorithm equals to the one obtained by Lin’s algorithm. At the same time, 25% of additions, 100% of multiplications, and 100% of exponents are reduced for computing the channel confidence. In addition, 1-db coding gain can be obtained by Lin's algorithm at the cost of the double of hardware complexity compared to Elia's algorithm.

Keywords: Soft decoding algorithm, Golay code, Field programmer gate array.

 

Received July 17, 2011; accepted May 22, 2012

  

Shubair Abdulla1, Sureswaran Ramadass2, Altyeb Altaher Altyeb3, and Amer Al-Nassiri4
1Instructional and Learning Technologies Department, Sultan Qaboos University, Oman 
2,3NAV6 Center of Excellence, Universiti Sains Malaysia, Malaysia
4IT College, Ajman University of Science and Technology, UAE

 

Abstract: We present a worm detection system that leverages the reliability of IP-Flow and the effectiveness of learning machines. Typically, a host infected by a scanning or an email worm initiates a significant amount of traffic that does not rely on DNS to translate names into numeric IP addresses. Based on this fact, we capture and classify NetFlow records to extract feature patterns for each PC on the network within a certain period of time. A feature pattern includes: no of DNS requests, no of DNS responses, no of DNS normals, and no of DNS anomalies. Two learning machines are used, K-Nearest Neighbors (KNN) and Naïve Bayes (NB), for the purpose of classification. Solid statistical tests, the cross-validation and paired t-test, are conducted to compare the individual performance between the KNN and NB algorithms. We used the classification accuracy, false alarm rates, and training time as metrics of performance to conclude which algorithm is superior to another. The data set used in training and testing the algorithms is created by using 18 real-life worm variants along with a big amount of benign flows.

Keywords: IP flow, netflow, NB, KNN, scanning worms, email worms.

 

Received September 27, 2011; accepted May 22, 2012

Faten Kharbat¹, Mohammed Odeh², Larry Bull^3
1Department of Management Information Systems, Al Ain University of Science and Technology, UAE
²Reader ＆ Leader of Software Engineering Research Group University of the West of England, UK
³Department of Computer Science and Creative Technologies, University of the West of England, UK

 

 

Abstract: This paper reports on the development of anew hybrid architecture that integrates Learning Classifier Systems (LCS) with Rete-based Production Systems Inference Engine to improve the performance of the process of compacting LCS generated rules. While LCS is responsible for generating a complete rule set from a given breast cancer pathological data-set, an adapted Rete-based inference engine has been integrated for the efficient extraction of a minimal and representative rule set from the original generated rule set. This has resulted in an architecture that is hybrid, efficient, component-based, elegant, and extensible. Also, this has demonstrated significant savings in computing the match phase when building on the two main features of the Rete match algorithm, namely structural similarity and temporal redundancy. Finally, this architecture may be considered as a new platform for research on compaction of LCS rules using Rete-based inference engines.

Keywords: Hybrid architecture, learning classifier systems, rete algorithm, production systems.

 

Received June 6, 2012; accepted February 12, 2013

  

Full Text

Marryam Murtaza, Muhammad Sharif, Mudassar Raza, and Jamal Hussain Shah
  Department of Computer Sciences,COMSATS Institute of Information Technology Wah Cantt, Pakistan

Abstract: Selecting a low dimensional feature subspace from thousands of features is a key phenomenon for optimal classification. Linear Discriminant Analysis (LDA) is a basic well recognized supervised classifier that is effectively employed for classification. However, two problems arise in intra class during Discriminant Analysis. Firstly, in training phase the number of samples in intra class is smaller than the dimensionality of the sample which makes LDA unstable. The other is high computational cost due to redundant and irrelevant data points in intra class. An Adaptive Margin Fisher’s Criterion Linear Discriminant Analysis (AMFC-LDA) is proposed that addresses these issues and overcomes the limitations of intra class problems. Small Sample Size problem is resolved through modified maximum margin criterion which is a form of customized LDA and Convex hull. Inter class is defined using LDA while intra class is formulated using quick hull respectively. Similarly, computational cost is reduced by reformulating within class scatter matrix through Minimum Redundancy Maximum Relevance (mRMR) algorithm while preserving Discriminant Information. The proposed algorithm reveals encouraging performance. Finally, a comparison is made with existing approaches.

Keywords: AMFC, LDA, Maximum margin criterion, quick hull, mRMR.

 

Received December 1, 2011; accepted May 22, 2012

  

Full Text