Accepted Papers

  • Bio-Inspired Metaheuristic based Visual Tracking and Ego-motion Estimation
    J.R.Siddiqui and Siamak Khatibi, Blekinge Institute of Technology, Sweden
    The problem of robust extraction of ego-motion from a sequence of images for an eye-in-hand camera configuration is addressed. A novel approach toward solving planar template based tracking is proposed which performs a non-linear image alignment and a planar similarity optimization to recover camera transformations from planar regions of a scene. The planar region tracking problem as a motion optimization problem is solved by maximizing the similarity among the planar regions of a scene. The optimization process employs an evolutionary metaheuristic approach in order to address the problem within a large on-linear search space. The proposed method is validated on image sequences with real as well as synthetic image datasets and found to be successful in recovering the ego-motion. A comparative analysis of the proposed method with various other state-of-art methods reveals that the algorithm succeeds in tracking the planar regions robustly and is comparable to the state-of-the art methods. Such an application of evolutionary metaheuristic in solving complex visual navigation problems can provide different perspective andcould help in improving already available methods..

  • Rule-Based Model Via Rough Sets to MIMO Discrete-Time Nonlinear Dynamical System
    Carlos A. M. Pinheiro and Rubiane H. Oliveira, Federal University of Itajuba (UNIFEI), Brazil
    This paper suggests a method to develop rule-based models to systems with multiple inputs and outputs using concepts about rough sets. The rules provide relations among variables and give a mechanism to link granular descriptions of the models with the computational procedures used. An estimation procedure is applied to compute values from numeric representations encoded by rule sets. The method is useful to develop models of nonlinear systems. An example of multi-input multi-output (MIMO) discrete-time nonlinear dynamical system is presented.

  • Optical Character Recognition System for Kashmiri Script
    Idrees Fazili,Shahid Ashraf Qadri and A. M. Wani,Central University of Kashmir,India
    This paper presents an approach towards the development and implementation of a basic OCR (Optical Character Recognition) system for printed text of Kashmiri language. The proposed system is a first attempt of its kind, towards the development of an OCR system for Kashmiri language. This system uses a segmentation based approach, in which an image is recursively segmented to the constituent character level. First the text image is segmented into lines of text then each of the lines of text is further segmented into words or ligatures, then words or ligatures are further decomposed into isolated characters. The next step is a character by character correlation for recognition of characters, whereby segmented character images are compared with a pre-defined reference set based on correlation coefficient, and corresponding codes are written into a text file; in the sequence the characters are found As the recognition process is completed, the character codes in the text file are given to the rendering engine, which displays the recognized text in a text region.

  • Intelligent Multi-Agent Fuzzy Control System Under Uncertainty
    Ben Khayut, Lina Fabri, Maya Abukhana, IDTS at Intelligence Decisions Technologies Systems,Israel
    The traditional control systems are a set of hardware and software infrastructure domain and qualified personnel to facilitate the functions of analysis, planning, decision-making, management and coordination of business processes. Human interaction with the components of these systems is done using a specified in advance script dialogue "menu", mainly based on human intellect and unproductive use of navigation. This approach doesn't lead to making qualitative decision and effective control, where the situations and processes cannot be structured in advance. Any dynamic changes in the controlled business process make it necessary to modify the script dialogue. This circumstance leads to a redesign of the components of the entire control system. In the autonomous Fuzzy Control System, where the situations are unknown in advance, fuzzy structured and artificial intelligence is crucial, the redesign described above is impossible. To solve this problem, we propose the data, information and knowledge based technology of creation Situational, Intelligent Multi-agent Control System, which interacts with users and/ or agent systems in natural and other languages, utilizing the principles of Situational Control and Fuzzy Logic theories, Artificial Intelligence, Linguistics, Knowledge Base technologies and others. The proposed technology is defined by a) methods of situational fuzzy control of data, information and knowledge, b) modelling of fuzzy logic inference, c) generalization and explanation of knowledge, d) fuzzy dialogue control, e) machine translation, f) fuzzy decision-making, g) planning and h) fuzzy control of organizational unit in real-time under uncertainty, fuzzy conditions, heterogeneous domains, multi-lingual communication in Fuzzy Environment.

  • The Applications of Data Mining in the Securities Market on R
    Lei Pang, Xingyu Chen, Beijing University of Posts and Telecommunications,China
    Did the fluctuation of stock price index just present a random Brownian motion or follow certain rules? There still is no consistent conclusion. In this paper, the rules and actions of price fluctuation are mainly explored and analyzed. Provided that the volatility of stock price follows a certain rule, we take the history quotations as research data to predict the transaction behaviors by model, and evaluate system performance in stock returns. The trading system combines with linear regression model, random forests, ANNs, SVM and MARS. Results show that there exist rules in the fluctuation of stock price index, the price in a certain period of future time will be affected by the current price, but the price can’t completely reflect the information in time. These methods and results to the real stock trading and theoretical analysis provide valuable reference and guidance

  • Robust Speech Recognition by Using Microphone Array
    Mohammadreza Seifikar,Polytechnic University of Turin,Italy
    The quality and performance of Automatic Speech Recognition algorithms (aka ASR) will reduce by increasing the distance, the noises, acoustic echoes and room reverberations. We are going to explore how an array of microphones can helps in order to improve the recognition quality and the out coming performances. This paper will explore, improve and extend the usability of ASR by means of analyzing and merging several techniques like BF, AEC, NR etc. considering a real environment, an array of 8 microphones could be applied on a TV set in order to reduce the echoes and focus on the user who is speaking and giving command to the TV. To improve flexibility and modularity we have followed a pipeline model combining several Plugins in order to reduce the noise, adjusting the gain, canceling the echo and finally recognizing the speech.

  • A Recommender System Sensitive to Intransitive Choice and Preference Reversals
    Amir Konigsberg and Ron Asherov, Princeton University, United States
    One of the basic foundations for many recommender systems is the assumption of preference consistency and transitivity of choice. In this paper we challenge this assumption and argue that it should be revised. We also provide a method by which recommender systems can estimate preference reversals and choice intransitivity. Our general approach is to incorporate variants of choice-behavior into our method such that the recommender system we propose incorporates decision-sensitive factors within choice-sets that tend to influence decision making.

  • Image Imputation Based on Clustering Similarity Comparison
    Sathit Prasomphan,King Mongkut’s University of Technology ,Thailand
    This paper presents a method to fill in missing data in an image. If the missing data are clustered in forms of an empty shape, then a similarity pattern searching and filling is performed.The missing data areas are divided into a set of windows of equal size. Each windowed area will be compared with every other non-missing data area of the original image to find the area that is most similar to the missing area. The experimental results show that in several cases our proposed algorithms outperform traditional methods.

  • Pre-Ranking Documents Valorization In The Information Retrieval Process
    Chkiwa Mounira, Jedidi Anis and Faiez Gargouri, Sfax University, Tunisia
    In this short paper we present three methods to valorise score relevance of some documents basing on their characteristics in order to enhance their ranking. Our framework is an information retrieval system dedicated to children. The valorisation methods aim to increase the relevance score of some documents by an additional value which is proportional to the number of multimedia objects included, the number of objects linked to the user particulars and the included topics. All of the three valorization methods use fuzzy rules to identify the valorization value.

  • An RBAC based Rule Generator and Inference Engine for Firewall
    Veena G, Amrita University, India
    The network security policy that describes the security requirements of an organization is presented in a high-level form. This high level policy is implemented using some low-level packet filtering rules, mainly firewall technology. One of the main difficulties faced by the network administrator is how to translate the high-level policy description to the low-level firewall rule-base.In this work, we use Role Based Access Control mechanism to manage the firewall security policies. The main concept of RBAC model is that the services are not directly assigned to the user, it is assigned to roles. A deductive database using Prolog is used to maintain the network policy of the organization. We propose a novel rule generator algorithm with linear complexity. This algorithm is used to generate low level packet filtering rules from the high-level network policy.

  • System Design of a Computer-Based Clinical Decision Support System Management by using of Bayeseian Network and Petri Net and Artificial Neural Network Approsch
    Neda Darvish1,Khikmat K.Muminov1 and Hoda Darvish2, 1S.U.Umarov of the Academy sciences,Iran , 2Islamic Azad university,Iran,
    Nowadays, the role of management science techniques to improve the quality of medical services is considerable. With effective management and good planning of human resources in a system of health care, improving the health status can be promoted. Hospital is one of the most well known organizations for providing these types of services that plays an important in preserving patients' health improvement. In this paper, we propose a novel method based on using of artificial intelligent networking for Medical – Management of System Design of Computer - Base for Re-Engineering for Hospital System. Patients and Methods: Using Bayeseian Network and Artificial Neural Network and Petri net for providing decision process in order to enhance the Hospital environment, medical management system modeling will be presented. The effective and optimal management of human resources in healthcare and treatment organization can be possible with appropriate evaluation of human resources units.

  • Fuzzy Inference System For Volt/Var Control In Distribution Substations In Isolated Power Systems
    Vega-Fuentes E, León-del Rosario S, Cerezo-Sánchez J M, Vega-Martínez A ,University of Las Palmas de Gran Canaria, Spain
    This paper presents a fuzzy inference system for voltage/reactive power control in distribution substations. The purpose is go forward to automation distribution and its implementation in isolated power systems where control capabilities are limited and it is common using the same applications as in continental power systems. This means that lot of functionalities do not apply and computational burden generates high response times. A fuzzy controller, with logic guidelines embedded based upon heuristic rules resulting from operators at dispatch control center past experience, has been designed. Working as an on-line tool, it has been tested under real conditions and it has managed the operation during a whole day in a distribution substation. Within the limits of control capabilities of the system, the controller maintained successfully an acceptable voltage profile, power factor values over 0,98 and it has ostensibly improved the performance given by an optimal power flow based automation system.

  • Neighborhood Ambiguity Driven Abstraction
    Bartosz Papis and Andrzej Pacut, Warsaw University of Technology , Poland
    State abstraction [1] is one of solutions to the curse of dimensionality [2] problem, and possibly allows real-life application of AI algorithms. We present a new state abstraction algorithm inspired by stimulus discrimination theory from behavioral psychology [3], [4] and by current work on bisimulation theory as applied to reinforcement learning[5], [6], [7]. The introduced algorithm demonstrates key issues in state space abstraction in discrete domain. Possible applications of this approach include finding abstractions for options [8] or MAXQ [9] hierarchies for high dimensional, continuous problems.

  • Query Proof Structure Caching for Incremental Evaluation of Tabled Prolog Programs
    Taher Ali1,Ziad Najem2 and Mohd Sapiyan1 , 1Gulf University for Science and Technology , Kuwait, 2Kuwait University,Kuwait
    The incremental evaluation of logic programs maintains the tabled answers in a complete and consistent form in response to the changes in the database of facts and rules. The critical challenges for the incremental evaluation are how to detect which table entries need to change, how to compute the changes and how to avoid the re-computation. In this paper we present an approach of maintaining one consolidate system to cache the query answers under the non-monotonic logic. We use the justificationbased truth-maintenance system to support the incremental evaluation of tabled PROLOG Programs. The approach used in this paper suits the logic based systems that depend on dynamic facts and rules to benefit in their performance from the idea of incremental evaluation of tabled Prolog programs. More precisely, our approach favors the dynamic rules based logic systems

  • Towards Universal Rating of Online Multimedia Content
    Lawrence Nderu1, Nicolas Jouandeau2, and Herman Akdag2,1Jomo Kenyatta University of Agriculture and Technology, Kenya,2University of Paris 8- LIASD ,France
    Most website classification systems have dealt with the question of classifying websites based on their content, design, usability, layout and such,few have considered website classification based on users’ experience. The growth of online marketing and advertisement has lead to fierce competition that has resulted in some websites using disguise ways so as to attract users.This may result in cases where a user visits a website and does not get the promised results. The results are a waste of time, energy and sometimes even money for users. In this context, we design an experiment that uses fuzzy linguistic model and data mining techniques to capture users’ experiences, we then use the k-means clustering algorithm to cluster websites based on a set of feature vectors from the users’ perspective. The content unity is defined as the distance between the real content and its keywords. We demonstrate the use of bisecting k-means algorithm for this task and demonstrate that the method can incrementally learn from user’s profile on their experience with these websites.

  • A ROS Implementation of the Mono-SLAM Algorithm
    Ludovico Russo1, Stefano Rosa1, Basilio Bona1 and Matteo Matteucci2,1Politecnico di Torino,Italy ,2Politecnico di Milano, Italy
    Computer vision approaches are increasingly used in mobile robotic systems, since they allow to obtain a very good representation of the environment by using low-power and cheap sensors. In particular it has been shown that they can compete with standard solutions based on laser range scanners when dealing with the problem of simultaneous localization and mapping (SLAM), where the robot has to explore anunknown environment while building a map of it and localizing in the same map. We present a package for simultaneous localization and mapping in ROS (Robot Operating System)using a monocular camera sensor only. Experimental results in real scenarios as well as on standard datasets show that the algorithm is able to track the trajectory of the robot and build a consistent map of small environments, while running in near real-time on a standard PC.

  • Design, Implement and Simulate an Agent Motion Planning Algorithm in 2D and 3D Environments
    Haissam El-Aawar and Hussein Bakri,Lebanese International University (LIU),Lebanon
    This article presents a computer simulated artificial intelligence (AI) agent that is able to move and interact in 2D and 3D environments. The agent has two operating modes: Manual Mode and Map or Autopilot mode. In the Manual mode the user has full control over the agent and can move it in all possible directions depending on the environment. In addition to that, the designed agent avoids hitting any obstacle by sensing them from a certain distance. The second and most important mode is the Map mode, in which the user can create a custom map, assign a starting and target location, and add predefined and sudden obstacles. The agent will then move to the target location by finding the shortest path avoiding any collision with any obstacle during the agent’s journey. The article suggests as a solution, an algorithm that can help the agent to find the shortest path to a predefined target location in a complex 3D environment, such as cities and mountains, avoiding all predefined and sudden obstacles. It also avoids these obstacles during manual control and moves the agent to a safe location automatically.

  • Genetic optimization for ransac matching used in 3D mapping of outdoor environments
    Boutine Rachid and Benmohammed Mouhamed, 1university Skikda, Algeria and 2Mentouri university Constantine, Algeria
    the navigation in unknown outdoor environments, require a set of high level skills like scene interpretation, and planning, and a set of basic skills like mapping , localization and obstacle avoidance. The robot when it moves in new environment should first, begin by discovering navigable space used later for path planning, and also discovering non navigable space used directly for obstacle avoidance and later in localization and mapping. In this paper, we have proposed a new features extraction algorithm, furthermore, used in the process of alignment of successive 3d point clouds collected by the 3d scanner of the robot. Also we have proposed a new matching technique for planar features, based on the combination of ransac and genetic algorithms, in order to finding the best alignment that minimize the global distance between all features, we have used the Bhattacharyya distance to measure the distance between 3d descriptors of high curvature points, for demonstrating the robustness of our proposed method, we shall testing it on real dataset collected by a Kurt3D robot. Lot of domains, can benefit from the results of this work, such as underground mines discovering, industrial automation, unmanned transportation, disaster rescue missions,…etc.

  • Auto Landing Process for Autonomous Flying Robot by Using Image Processing Based on Edge Detection
    Bahram Lavi Sefidgari and Sahand Pourhassan Shamchi, EMU, Cyprus
    In today’s technological life, everyone is quite familiar with the importance of security measures in our lives. So in this regard, many attempts have been made by researchers and one of them is flying robots technology. One well-known usage of flying robot, perhaps, is its capability in security and care measurements which made this device extremely practical, not only for its unmanned movement, but also for the unique manoeuvre during flight over the arbitrary areas. In this research, the automatic landing of a flying robot is discussed. The system is based on the frequent interruptions that is sent from main microcontroller to camera module in order to take images; these images have been distinguished by image processing system based on edge detection, after analysing the image the system can tell whether or not to land on the ground. This method shows better performance in terms of precision as well as experimentally.