Georgios Miaoulis and Dimitri Plemenos (Eds.) Intelligent Scene Modelling Information Systems

Transcription

1 Georgios Miaoulis and Dimitri Plemenos (Eds.) Intelligent Scene Modelling Information Systems

2 Studies in Computational Intelligence, Volume 181 Editor-in-Chief Prof. Janusz Kacprzyk Systems Research Institute Polish Academy of Sciences ul. Newelska Warsaw Poland kacprzyk@ibspan.waw.pl Further volumes of this series can be found on our homepage: springer.com Vol Marina Gavrilova (Ed.) Generalized Voronoi Diagram: A Geometry-Based Approach to Computational Intelligence, 2009 ISBN Vol Dimitri Plemenos and Georgios Miaoulis (Eds.) Artificial Intelligence Techniques for Computer Graphics, 2009 ISBN Vol P. Rajasekaran and Vasantha Kalyani David Pattern Recognition using Neural and Functional Networks, 2009 ISBN Vol Francisco Baptista Pereira and Jorge Tavares (Eds.) Bio-inspired Algorithms for the Vehicle Routing Problem, 2009 ISBN Vol Costin Badica, Giuseppe Mangioni, Vincenza Carchiolo and Dumitru Dan Burdescu (Eds.) Intelligent Distributed Computing, Systems and Applications, 2008 ISBN Vol Pawel Delimata, Mikhail Ju. Moshkov, Andrzej Skowron and Zbigniew Suraj Inhibitory Rules in Data Analysis, 2009 ISBN Vol Nadia Nedjah, Luiza de Macedo Mourelle, JanuszKacprzyk,FelipeM.G.França and Alberto Ferreira de Souza (Eds.) Intelligent Text Categorization and Clustering, 2009 ISBN Vol Djamel A. Zighed, Shusaku Tsumoto, Zbigniew W. Ras and Hakim Hacid (Eds.) Mining Complex Data, 2009 ISBN Vol Constantinos Koutsojannis and Spiros Sirmakessis (Eds.) Tools and Applications with Artificial Intelligence, 2009 ISBN Vol Ngoc Thanh Nguyen and Lakhmi C. Jain (Eds.) Intelligent Agents in the Evolution of Web and Applications, 2009 ISBN Vol Andreas Tolk and Lakhmi C. Jain (Eds.) Complex Systems in Knowledge-based Environments: Theory, Models and Applications, 2009 ISBN Vol Nadia Nedjah, Luiza de Macedo Mourelle and Janusz Kacprzyk (Eds.) Innovative Applications in Data Mining, 2009 ISBN Vol Lakhmi C. Jain and Ngoc Thanh Nguyen (Eds.) Knowledge Processing and Decision Making in Agent-Based Systems, 2009 ISBN Vol Chi-Keong Goh,Yew-Soon Ong and Kay Chen Tan (Eds.) Multi-Objective Memetic Algorithms, 2009 ISBN Vol I-Hsien Ting and Hui-Ju Wu (Eds.) Web Mining Applications in E-Commerce and E-Services, 2009 ISBN Vol Tobias Grosche Computational Intelligence in Integrated Airline Scheduling, 2009 ISBN Vol. 174.Ajith Abraham, Rafael Falcón and Rafael Bello (Eds.) Rough Set Theory: A True Landmark in Data Analysis, 2009 ISBN Vol Godfrey C. Onwubolu and Donald Davendra (Eds.) Differential Evolution: A Handbook for Global Permutation-Based Combinatorial Optimization, 2009 ISBN Vol Beniamino Murgante, Giuseppe Borruso and Alessandra Lapucci (Eds.) Geocomputation and Urban Planning, 2009 ISBN Vol Dikai Liu, Lingfeng Wang and Kay Chen Tan (Eds.) Design and Control of Intelligent Robotic Systems, 2009 ISBN Vol Swagatam Das, Ajith Abraham, Amit Konar Metaheuristic Clustering, 2009 ISBN Vol Mircea Gh. Negoita, Sorin Hintea Bio-Inspired Technologies for the Hardware of Adaptive Systems, 2009 ISBN Vol Wojciech Mitkowski, Janusz Kacprzyk (Eds.) Modelling Dynamics in Processes and Systems, 2009 ISBN Vol Georgios Miaoulis, Dimitri Plemenos (Eds.) Intelligent Scene Modelling Information Systems, 2009 ISBN

3 Georgios Miaoulis Dimitri Plemenos (Eds.) Intelligent Scene Modelling Information Systems 123

4 Georgios Miaoulis Technological Education Institution of Athens Dept. Computer Science Ag. Spyridonos Str Athens Egaleo Greece Dimitri Plemenos Universite Limoges Laboratoire MSI 83 rue d Isle Limoges France plemenos@unilim.fr ISBN e-isbn DOI / Studies in Computational Intelligence ISSN X Library of Congress Control Number: c 2009 Springer-Verlag Berlin Heidelberg This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Typeset & Cover Design: Scientific Publishing Services Pvt. Ltd., Chennai, India. Printed in acid-free paper springer.com

5 Preface Scene modeling is a very important part in Computer Graphics because it allows creating more or less complex models to be rendered, coming from the real world or from the designer s imagination. However, scene modeling is a very difficult task, as there is a need of more and more complex scenes and traditional geometric modelers are not well adapted to computer aided design. Even if traditional scene modelers offer very interesting tools to facilitate the designer s work, they suffer from a very important drawback, the lack of flexibility, which does not authorize the designer to use incomplete or imprecise descriptions, in order to express his (her) mental image of the scene to be designed. Thus, with most of the current geometric modelers the user must have a quite precise idea of the scene to design before using the modeler to achieve the modeling task. This kind of design is not really a computer aided one, because the main creative ideas have been elaborated without any help of the modeler. Declarative scene modeling could be an interesting alternative to traditional geometric modeling. Indeed, declarative scene modeling tries to give intuitive solutions to the scene modeling problem by using Artificial Intelligence techniques which allow the user to describe high level properties of a scene and the modeler to give all the solutions corresponding to imprecise properties. No matter if a scene modeler is declarative or a classical geometric one, it is usually not enough to help the user in the designing process. The designer has not always to start from nothing and to design a new scene. Scene data bases, or scene knowledge bases in the case of declarative modeling, should be associated to the scene modeler, in order to allow the designer to use and/or to modify existing models during the designing process. Moreover, easy access to Internet, together with appropriate tools, should be available for the user, allowing, for example, collaborative scene modeling. In other words, scene modeling tools should be complete information systems, that is scene modeling information systems. This book is dedicated to intelligent scene modeling information systems, that is information systems using Artificial Intelligence techniques to design scenes. Declarative scene modeling techniques are presented, as well as their implementation in an intelligent information system. In order to improve efficiency of declarative modeling based scene modeling information systems, various techniques are proposed in the book: coupling of a de-

6 VI Preface clarative modeler with a classical geometric modeler. Use of machine-learning Artificial Intelligent techniques, allowing the system to learn the user preferences; introduction of high level concepts, especially the concept of style in architectural design; introduction of collaborative declarative modeling techniques. The last chapter is dedicated to the web security problem, as an intelligent scene modeling information system must be able to work in open environments. Artificial Intelligence techniques, as well as information visualization techniques, are proposed to easily detect web attacks. Intelligent scene modeling information systems may reduce the complexity of scene modeling, by using well adapted Artificial Intelligence techniques, together with a lot of libraries and other tools, allowing easy description, search or modification of scenes. The use of declarative modeling techniques in the heart of such systems was experimented and seems to give interesting results, even if these techniques have to be still improved. Georgios Miaoulis Dimitri Plemenos

7 Contents 1 Intelligent Scene Modelling Information Systems: The Case of Declarative Design Support... 1 Georgios Miaoulis 1.1 Introduction The Scene Modelling Process in Declarative Design Support Information, Knowledge and Scene Models Representations Physical Scene Models Conceptual Scene Models Generic Models Scene Conceptual Modelling in MultiCAD Software Architectures for Declarative Design Support MultiCAD: Objectives, Constraints and Functional Choices Definition of MultiCAD Framework-Architecture Conclusion References Declarative Modeling in Computer Graphics Dimitri Plemenos 2.1 Introduction What Is Declarative Scene Modeling Imprecision Management in Declarative Modelers A Classification of Declarative Scene Modelers Modelers Using Exploration Mode in Scene Generation Modelers Using Solution Search Mode in Scene Generation Other Declarative or Declarative-Like Modelers Scene Understanding in Declarative Scene Modeling Constraint Satisfaction Techniques for Declarative Scene Modeing Arithmetic Constraint Satisfaction Techniques The Resolution Process Constraint Logic Programming on Finite Domains CLP(FD) Hierarchical Decomposition-Based Improvements Geometric Constraint Satisfaction Techniques Principles of the MultiFormes Geometric Constraint Solver

8 VIII Contents The Resolution Process The Intersection and Sampling Problems Some Other Problems Discussion Arithmetic CSP Geometric CSP Declarative Scene Modeling and Machine-Learning Techniques A Dynamical Neural Network for Filtering Unsatisfactory 49 Solutions in DMHD Structure of the Used Network The Machine Learning Process Discussion Advantages and Drawbacks of Declarative Scene Modeling Future Issues Conclusion References Understanding Scenes Vassilios S. Golfinopoulos 3.1 Introduction to Reverse Engineering Reverse Engineering in Scene Modelling Reverse Engineering and Geometric Modelling Reverse Engineering and Feature-Based Modelling Reverse Engineering and Declarative Modelling Integration of the Two Models Reconstruction Phase Extended Design Methodology System Architecture Data and Knowledge Storage The Stratified Representation Extraction of Relations and Properties Scene Modifications The Propagation Policy The Unified Stratified Representation The Resultant Declarative Description Conclusions References Intelligent Personalization in a Scene Modeling Environment Georgios Bardis 4.1 Introduction Intelligent Personalization and Contributing Fields Preference Model Preference Structure User Preference as a Function Multicriteria Decision Support... 95

9 Contents IX Outranking Methods Weighted Sum Methodologies Machine Learning Traditional Machine Learning Mechanisms Incremental Learning Imbalanced Datasets Context Specific Issues Intelligent Personalization in a Scene Modeling Environment Scene Representations Scene Modeling Process Solution Generation: Constraint Satisfaction 104 Techniques Solution Generation: Evolutionary Techniques Solution Visualization Scene Modeling Environment Preferences Acquisition Solution Encoding for Preferences Acquisition Preferences Acquisition via Incremental Learning User-Assisted Acquisition of Preferences Intelligent User Profile Module Architecture Declarative Modeling Module Architecture Experimental Results Performance Indices and Representative Scenes Experiment Series Conclusion References Web-Based Collaborative System for Scene Modelling 121 John Dragonas, Nikolaos Doulamis 5.1 Introduction Research Scope Related Work Collaborative Design Collaborative Systems Declarative Design Overview of MultiCAD Architecture DKABM Framework Declarative Design Representations Collaborative Declarative Modelling System Web-Based CDMS Framework Declarative Collaborative Module Case Study Study of Collaborative Activity Team Profile Module Single Designer Approach Intelligent Profile Estimation

10 X Contents Recursive Implementation Collaborative Approach Preference Consensus Module Collaborative Clustering Simulations Conclusions References Aesthetic Aided Intelligent 3D Scene Synthesis. 153 Dimitrios Makris 6.1 Introduction Research Scope Proposed Methodology Contributing Areas Related Work Evolutionary Computing Techniques Evolutionary Design Genetic Algorithm Applications in Design Computational Aesthetic Approaches Style Modelling Approaches The Concept of Style MultiCAD Framework Style Research Approach Architectural Style Modelling Style Knowledge Framework Measure of Style Multi-objective Genetic Algorithm Genetic Algorithm MOGA Mechanism Implementation Framework Software Architecture User Interface Layer Processing Layer Data Management Layer System Evaluation Discussion Conclusions Declarative Modelling and Architectural Conceptual 179 Design Aesthetic and Artificial Intelligence References Network Security Surveillance Aid Using Intelligent Visualization for Knowledge Extraction and Decision Making Ioannis Xydas 7.1 Introduction Web Security Intrusion Detection

11 Contents XI Visualization Visual Data Analysis Research Objectives Related Work Visualization Prototype System Data Capture Module Pre-processor Module Knowledge Base Module Classes of Web Attacks Training Data Quality Evolutionary Artificial Neural Network EANN Performance Versus ANN Graph Generator Module Statistical Analysis Module Prototype System Performance Introduction Classification Neyman-Pearson Decision Rule Sufficient Statistics and Monotonic Transformations Neyman-Pearson Lemma: General Case Detection, False and Miss Probabilities of the Prototype 208 System ROC Curve of the Prototype System Conclusion References Author Index

12