In the last ten years, skin detection has been a milestone in most of the computer vision applications. But till now, there is no robust skin detector. The different degrees of the skin tone color are the obstacle that faces skin detection process. This paper proposes an adaptive skin modeling and detection technique which is based on face skin tone color. Face is a good indicator of different characteristics of skin tone color where it carries significant information about skin color. Skin modeling aims to develop adaptive margins of skin detector. These margins have been obtained after applying an online dynamic threshold to the pixels gathered around the major and minor axes of bounding rectangle of detected face. Experimental results show that the proposed method has promising results compared to state-of-the-art skin detection methods.
Fuzzy optimization models provide a powerful decision support tool for optimization models in fuzzy environment. In this paper fuzzy goal programming (FGP) is integrated with the fuzzy analytic hierarchy process (FAHP) to determine optimal plant and distribution centre locations in a supply chain with special focus on the operational efficiencies of the distribution centres. The integrated FGP-FAHP model incorporates multiple conflicting objectives as demanded by the decision process. The concept of fuzzy logic is utilized to model the variation of demands at retails centres that makes the model more sophisticated to the real SCM problem. The FAHP is used to model the decision maker preferences and to handle information ambiguaties in the comparison judgment by introducing a linguistic variable, and to find the relative weights of multiple objectives in FGP. In addition to, the proposed FGP-FAHP provides a risk management decision support tool in SCM problem associated with information ambiguities.
Person Name extraction from Arabic text is a challenging task. While most existing Arabic texts are written in Modern Standard Arabic Text (MSA) the volume of Arabic Colloquial text is increasing progressively with the wide spread use of social media examples of which are Facebook, Google Moderator and Twitter. Previous work addressed extracting persons' names from MSA text only and especially from news articles. Previous work also relied on a lot of resources such as gazetteers for places, organizations, verbs, and person names. In this paper we introduce a system for extracting persons' names from any type of Arabic text whether it is MSA or Colloquial using very few resources. In our system, Natural Language Processing (NLP) is integrated with a limited set of dictionaries to extract a person's name from Arabic text. The paper also presents the results of evaluating the system on two datasets, one for MSA and the other for Colloquial Arabic. The results achieved were found to be satisfactory in terms of precision, recall and f-measure.
Most current implementations of quantum key distribution (QKD) are point-to-point systems with one sender transmitting to only one receiver. Development of these single-receiver systems has now reached a comparatively advanced stage. However, many communication systems operate in a point-to- multi-point (multiple-receiver) configuration rather than in point-to-point mode so it is crucial to demonstrate compatibility with this type of network in order to maximize the application range for QKD. In this paper, we suggest a proposed architecture of Point-to-Multipoint QKD (QKDP2MP) Systems and bow it will present in terms of QBB and Quantum layered Architecture.
Filling pathway hole is a point of research in the field of Bioinformatics especially in metabolic pathway where the analysis of metabolic pathways is an essential topic in understanding the relationship between genotype and phenotype [4]. The pillar of the research cycle is the data collection which precedes the analysis phase to solve the pathway hole. The required data for this area is scattered among different data sources, which represent a problem for the researchers of this area. This paper provides a solution to this obstacle by collecting the required data from various data sources in one database RGB MAPS. Also we have developed a tool that could be used by other researchers to analyze the pathway holes.
Software architecture is a key discipline in software engineering as it performs a central role in many modern software development paradigms. For an evolving complex architecture, assessing the change impact for the components considering all maintenance scenarios is a difficult problem. In this paper, we present a methodology to conduct sensitivity analysis of maintainability-based risk factors for software architectures. The methodology can assist the software architect to determine the components with the least change impact. Two case studies are used to illustrate the methodology. Results show that only small subset of components are highly sensitive to change.
This paper proposes a decision support model to determine the contribution of urban region in the National Target Total Fertility Rate using Fuzzy Analytical Hierarchical Process (AHP). The proposed model supports in regional sustainability developing regarding family planning programs by determining, and reducing the contribution of each governorate in Urban region of TFR and its fertility rate taking into consideration its socio-economic context including women education level, percent of young women, infant mortality, age at first marriage, women working status, spouse level of education, family income, and place of residence. The FAHP is employed to determine the contribution of each governorate in urban region to the required reduction of regional TFR targets. All assessments of selected criteria are assessed and because of uncertainty we are using fuzzy AHP by five experts' opinions in the domain of family planning to reflect its impact and contributions into governorate TFR. The proposed model is implemented in urban region of Egypt to disaggregate the national TFR into regional TFR to achieve its required reduction level. The obtained results showed that the proposed model performed well in introducing a set of governorates targets that achieves the regional goal and is consistent with the characteristics of each governorate.
In this paper, a parallel iterative finite difference method (PIFD) for solving 2D Poisson's equation on a distributed system using Message Passing Interface (MPI) is investigated. This method is based on the domain decomposition method, where the 2D domain is divided into multiple sub-domains using horizontal and/or vertical axis depending on the available number of computer nodes. For interior points Poisson's equation is solved implicitly by four iterative schemes in combining with the boundary conditions. At the interface points of interior subdomains, Poisson's equation is solved by explicit iterative schemes. The proposed approach fulfills the suitability for the implementation on Linux PC cluster through the minimization of inter-process communication by restricting the exchange of data to the interface between the sub-domains. To examine the efficiency and accuracy of the iterative algorithm, several numerical experiments using different number of nodes of the Linux PC cluster are tested. The performance metrics clearly show the benefit of using the proposed approach on the Linux PC cluster in terms of execution time reduction and speedup with respect to the sequential running in a single PC.
This paper presents an optimized approach for mining opinions in Arabic Religious Decrees using an improved “Semantic Orientation using Pointwise Mutual Information” Algorithm. The original approach executed a number of steps to classify a religious decree into either Halal (Allowed) or Haraam (Prohibited). Those steps included Data Collection, Simple Text Preprocessing, Manual Data labeling, Advanced Text Preprocessing, Weight Calculation and experimentation using Supervised and Unsupervised Learning Algorithms. Weight Calculation process utilized SO-PMI Algorithm proposed by Wang and Araki in 2008. Results obtained by original approach gave an accuracy rate of 73.08%. The new approach utilizes an improved SO-PMI Algorithm that executes a series of advanced steps to improve the calculation of the weights. The improved algorithm increased the accuracy rate of the Unsupervised Learning Algorithm up to 20% but produced poor results for the Supervised Learning Algorithm.
The growing productions of maps are generating huge volumes of data that exceed people's capacity to analyze them moreover these data sets have different resources and types. It seems appropriate to apply knowledge discovery methods like data mining to spatial data so, one of the most significant application in spatial data mining is classification for remote sensing images. This paper proposes a classification system for remote sensing ASTER satellite imagery using SVM with non-linear kernel functions. The proposed system starts with the identification of selected area of study. This is followed by a preprocessing phase to enhance the quality of the input remote sensing satellite image and to reduce speckle without destroying the important features using mapping polynomial algorithm as geometric correction. Followed by, applying threshold algorithm for image segmentation. Then features are extracted using object based algorithm. Followed by, image classification using SVM with nonlinear kernel function. It is tested and evaluated on selected area of interest in the north-eastern part of the Eastern Desert of Egypt (Halaib Triangle). The obtained results carried out that SVM with RBF kernel function has the highest classification accuracy ratio.
The document that should appear here is not currently available.
Identifying moving objects from a video sequence is a fundamental and critical talk in many computer-vision application. It is an Important part of visual tracking system. In this paper, a new algorithm for background subtraction has been proposed and its performance wat investigated and compared. It is a combination of three background subtraction algorithms; frame difference, approximated median, and Mixture of Ganssian. Each algorithm first they are modified and then a decision level fusion algorithm applied to the result. The performance of these algorithms are compared before and after combinations. It has been found that the percentage error for the pixel in each frame has reduced remarkably when applying the combinational algorithm.
Electroencephalography (EEG) is the recording of electrical activity occurring in the brain, which is recorded from the scalp through placement of voltage sensitive electrodes. It has been repeatedly demonstrated that the brain emits voltage fluctuations on a continuous basis. These fluctuations are a reflection of the on-going brain dynamics, which present as a series of fluctuations that have characteristic waveforms and amplitude patterns, depending on the cognitive state of the subject. A number of published reports have indicated that there is enough depth in the EEG recording, rendering it suitable as a tool for person authentication. This idea has a solid underpinning in that recent evidence suggests much of the on-going EEG recordable activity within brains has a genetic component. This study presents the common steps for developing a human identification systems based on EEG signals. It will also present some of the important techniques used.
The complexity of software projects is increasing rapidly and in turn both cost and time of the testing process have become a major proportion of the software development process. Thus, it has become a strong motivation to investigate model based testing methods to decrease the overall cost, effort and time of the testing process by automating the generation of the test cases as well as their execution. This paper proposes an enhanced approach for automatically generating test cases from activity diagrams. Category partition method is applied to generate the final set of reduced test cases. The proposed model validates the generated test paths during the generation process to ensure that they meet a hybrid coverage criterion. The proposed model is automated and applied to around forty different case studies in different domains. Experimental evaluation is demonstrated to prove that the proposed model saves time and cost, thus increases the performance of the testing process.
One of the several benefits of text classification is to automatically assign document in predefined category. Researchers using LVQ algorithm in English and Persian [1, 2] and don't be attention for Arabic language. So in our research, we used neural network approach for classify Arabic text by using Learning Vector Quantization (LVQ) algorithm. This algorithm is based on Kohonen self organizing map (SOM) that is able to organize big-size document collections according to textual similarities. Also, LVQ algorithm requires less training examples and its faster than other classification methods. We select Arabic documents from different domains. After that we select suitable preprocessing methods such as term weighting schemes, and Arabic morphological analysis (stemming and light stemming), these preprocessing prepared dataset that need for classification. Then, we compared the results obtained from different LVQ improvement versions (LVQ2.1, LVQ3, OLVQ1 and OLVQ3). The results showed that the LVQ's algorithms especially LVQ2.1 algorithm achieved high accuracy and less time compared to other LVQ's algorithms.
Call center performance could be measured according to many metrics which can be expressed by a single value only called a key performance indicator (KPI). These indicators may differ from system to system according to its nature. Basically, call centers can be characterized as stochastic systems that could be modeled mathematically using queueing models (QM). This paper introduces a methodology for evaluating call centers performance using queueing models (QM) with and without customer abandonment, and studies the impact of abandonment rate as a metric of evaluating call center performance. In addition to, introduce a framework of the proposed methodology and a case study for three models. The propose methodology is implemented and analyzed on real benchmark datasets.
Virtual Enterprise (VE) is a temporary alliance of separate enterprises created to act together to share core competencies and resources in order to respond to market requirements of high quality, low cost, customer satisfaction, and quick responsiveness. The composition of virtual enterprises, which includes the selection of partners, is a vital issue that affects the success of VEs. This paper focuses on the solution procedure of the multi-objective partner selection problem in virtual enterprises where the cost coefficients are expressed as interval values by the decision maker. The paper uses a multi-objective algorithm, namely Pareto Simulated Annealing (PSA), and results showed improved values of the Pareto set as the iterations advances to get closer to each other until forming the final well-distributed Pareto front.
An enhanced predication approach for the network dropped packets problem is introduced. This work along with test results shows the possibility of guessing when the dropped packet occurs and also the source address for it. Since artificial neural networks (ANNs) algorithms are able to model nonlinear relations between different data sets, a proposed ANN based on particle swarm optimization training algorithm (PSO) is proposed. This global optimization algorithm is applied to the proposed ANN to avoid the local minima problem in the gradient descent-training algorithm and to achieve acceptable solution. The Particle swarm optimization technique is used in this work to optimize the performance of radial basis function artificial neural network (RBF-ANN). The data used in training and testing is the data collected by a particular network simulator (NS-2) program which is utilized to simulate the data for the neural network. This RBF-ANN model has been verified by comparing ANN simulated and test data. The presented results are obtained through the use of MATLAB 8.5 software from Math works.
The main goal of Next génération network networks is to provide to men a continuous connectivity with better Quality of Service (QoS). Or, rame problems con generate network's performances degradation like congestion state. In this paper, we propose a new congestion control mechanism in the radio Interface of the wireless heterogeneous networks, specifically in case of interworking Mesh network (IEEE 802.11s) with the cellular network 3gpp LTE. At the end, performance of the new congestion control scheme has been studied through simulation.
The proportional-integral-derivative (PID) controllers are the most popular controllers used in industry because of their remarkable effectiveness, simplicity of implementation and broad applicability. This paper presents an artificial intelligence (AI) method of particle swarm optimization (PSO) algorithm for tuning the optimal PID controller parameters for industrial process. PSO is a technique used to explore the search space of a given problem to find the settings or parameters required to maximize a particular objective. The focus will be on the application of the PSO into one of the popular problem setups in the engineering application area of control systems, which is called the inverted pendulum. This approach has superior features, including easy implementation, stable convergence characteristic and good computational efficiency over the conventional methods. The controller is obtained and validated by simulation; it's implemented to control the pendulum-cart system.
