Volume 2, Issue 3, 2012

By | September 4, 2018

 Performance Comparison of Face Recognition using Transform Domain Techniques

Jossy P. George, Department of Computer Science, Christ University, Bangalore, India.
Saleem S Tevaramani, K B Raja
Electronics and Communication Engineering Department
University Visvesvaraya College of Engineering, Bangalore, India.

Abstract: The biometrics is a powerful tool to authenticate a person for multiple applications. The face recognition is better biometrics compared to other biometric traits as the image can be captured without the knowledge and cooperation of a person. In this paper, we propose Performance Comparison of Face Recognition using Transform Domain Techniques (PCFTD). The face databases L – Spacek, JAFFE and NIR are considered. The features of face are generated using wavelet families such as Haar, Symelt and DB1 by considering approximation band only. The face features are also generated using magnitudes of FFTs. The test image features are compared with database features using Euclidian Distance (ED). The performance parameters such as FAR, FRR, TSR and EER computed using wavelet families and FFT. It is observed that the performance of FFT is better compared to wavelet families. The success rate of recognition is 100% for L – Spacek and JAFFE face databases as compared to 95% for NIR face databases

Keywords : Face Recognition; DWT; FFT; ED; Biometrics.

Leader Election Algorithm in 3D Torus Networks with the Presence of One Link Failure

Mohammed Refai, Department of SE, Faculty of Science and Information Technology, Zarqa University, Jordan.
Ibrahim AL-Oqily, Department of CS, Faculty of Prince Al-Hussein Bin Abdullah II, for Information Technology Hashemite University, Jordan.
Abed Alhamori, Faculty of Information Technology Albalqa University, Jordan.

Abstract: Leader election is the process of choosing a leader for symmetry breaking where each node in the network eventually decides whether it is a leader or not. This paper proposes a new leader election algorithm to solve the problem of leader failure in three dimensional torus networks. The proposed algorithm solves the election problem despite the existent of link failure. In a network of N nodes connected by three dimensional torus network, the new algorithm needs O(N) messages to elect a new leader in time steps. These results are valid for two cases: the simple case where the leader failure is detected by one node, and the worst case where the failure is discovered by N-1 nodes.

Keywords : Concurrency; Leader Election; Link Failure; leader failure; 3D Torus Networks.

A Three Stages Segmentation Model for a Higher Accurate off-line Arabic Handwriting Recognition

Said Elaiwat, School of Computer Science & Software Engineering (CSSE), The University of Western Australia (UWA), Australia 35Stirling Highway CRAWLEY WA 6009 University.
Marwan AL-abed Abu-zanona, Department of Computer Science , Imam Muhammad Ibn Saud Islamic, Al-Ehsa Branch, Al-Ehsa – Saudi Arabia.
Farah Hanna AL-Zawaideh, Department of Computer Information System, Irbid National University, Irbid, Jordan.

Abstract: Arabic handwriting recognition considers a one of the hardest applications of OCR system. The reason of that relates to characteristics of Arabic characters and the way of writing cursively. Furthermore, no rules can control on handwriting way, different styles, sizes and curves make the process of recognition is very complex. On other side, the key for reaching to good recognition is by getting a correct segmentation. Actually, the way of segmentation is important, because if there is a small part is not clear in character that will reflect on recognition process. In this paper we aim to enhance the accuracy of off-line Arabic Hand Written text segmentation. Three stages are proposed to reach to highest ratio of segmentation. Line segmentation is the first stage, where it is proposed to separate each line. We depend on row density to predict spaces among lines. Second stage is Object segmentation and it is proposed to segment each word or sub word. Eight neighbors connectivity are used to detect connected pixels. Final stage is shape segmentation which is proposed to segment sub word to characters. The idea in this stage is finding segmentation points among branch points in the baseline. To apply that we propose four threshold values to investigate on each branch point. The result was satisfactory and the model proved a good ability to tackle different types of texts with bad samples.

Keywords : Arabic handwritten recognition; Segmentation; Image processing; Pattern recognition.

Learning Vector Quantization (LVQ) and k-Nearest Neighbor for Intrusion Classification

Reyadh Shaker Naoum, Zainab Namh Al-Sultani
Department of Computer Science, Faculty of Information Technology, Middle East University, Amman, Jordan.

Abstract: Attacks on computer infrastructure are becoming an increasingly serious problem nowadays, and with the rapid expansion of computer networks during the past decade, computer security has become a crucial issue for protecting systems against threats, such as intrusions. Intrusion detection is an interesting approach that could be used to improve the security of network system. Different soft-computing based methods have been proposed in recent years for the development of intrusion detection systems. This paper presents a composition of Learning Vector Quantization artificial neural network and k-Nearest Neighbor approach to detect intrusion. A Supervised Learning Vector Quantization (LVQ) was trained for the intrusion detection system; it consists of two layers with two different transfer functions, competitive and linear. Competitive (hidden) and output layers contain a specific number of neurons which are the sub attack types and the main attack types respectively. k-Nearest Neighbor (kNN) as a machine learning algorithm was implemented using different distance measures and different k values, but the results demonstrates that using the first norm instead the second norm and using k=1 gave the best results among other possibilities. The experiments and evaluations of the proposed method have been performed using the NSL-KDD 99 intrusion detection dataset. Hybrid (LVQ_kNN) was able to classify the datasets into five classes at learning rate 0.09 using 23 hidden neurons with classification rate about 89%.

Keywords : Intrusion Detection System; Learning Vector Quantization; k-Nearest Neighbor.

Performance Evaluation for VOIP over IP and MPLS

Reyadh Shaker Naoum, Mohanand Maswady
Computer Information System Department, Faculty of Information Technology, Middle East University, Amman, Jordan.

Abstract:Corporates and multisite organizations are now applying VOIP usage all over their branches, this made offices with no boundaries and reduced a huge amount of cost for their infrastructure; facilitated exchanging for voice, video and Data .Growing demand for such usage has pushed the wheel for improving and applying more techniques to make this service more reliable, efficient and scalable. In this paper a simulation were performed and compared for a multisite office network for G.723 VOIP communication traffic applied on two network infrastructure models: one for IP and the other for MPLS, the results came encouraging for the MPLS model.

Keywords- component; MPLS; VOIP; CODECS; Multisite Offices.


Data set property based ‘K’ in VDBSCAN Clustering Algorithm

Abu Wahid Md. Masud Parvez
Software Quality Architect of Software quality department , Tech Prolusion Labs, San Francisco, USA.

Abstract— The term cluster analysis (first used by Tryon, 1939) encompasses a number of different algorithms and methods for grouping objects of similar kind into respective categories. Among different types of cluster the density cluster has advantages as its clusters are easy to understand and it does not limit itself to shapes of clusters. But existing density-based algorithms are lagging behind. The main drawback of traditional clustering algorithm which was largely recovered by VDBSCAN algorithm. But in VDBSCAN algorithm the value of parameter ‘K’ which was a user input dependent parameter. It largely degrades the efficiency of permanent Eps. In our proposed method the Eps is determined by the value of ‘k’ in varied density based spatial cluster analysis by declaring ‘k’ as variable one by using algorithmic average determination and distance measurement by Cartesian method and Cartesian product on multi dimensional spatial dataset where data are sparsely distributed. The basic idea of calculated ‘k’ which is computed from the characteristics of the examining dataset instead of a static user dependent parameter for increasing the efficiency of the VDBSCAN cluster analysis algorithm. By calculating value of ‘k’ with our newly developed arithmetic and algebraic method, user will obtain the most optimal value of Eps for
determining cluster for the sparsely distributed dataset. This will add significant amount of efficiency of the VDBSCAN cluster analysis algorithm.

Keywords— Data mining; Cluster analysis; Clustering algorithm; DBSCAN algorithm & Ep.

Leave a Reply

Your email address will not be published. Required fields are marked *