123

Smart Innovation, Systems and Technologies 55

Giuseppe De PietroLuigi GalloRobert J. HowlettLakhmi C. Jain Editors

Intelligent Interactive Multimedia Systems and Services 2016

Smart Innovation, Systems and Technologies

Volume 55

Series editors

Robert James Howlett, KES International, Shoreham-by-sea, UKe-mail: rjhowlett@kesinternational.org

Lakhmi C. Jain, University of Canberra, Canberra, Australia;Bournemouth University, UK;KES International, UKe-mails: jainlc2002@yahoo.co.uk; Lakhmi.Jain@canberra.edu.au

About this Series

The Smart Innovation, Systems and Technologies book series encompasses thetopics of knowledge, intelligence, innovation and sustainability. The aim of theseries is to make available a platform for the publication of books on all aspects ofsingle and multi-disciplinary research on these themes in order to make the latestresults available in a readily-accessible form. Volumes on interdisciplinary researchcombining two or more of these areas is particularly sought.

The series covers systems and paradigms that employ knowledge andintelligence in a broad sense. Its scope is systems having embedded knowledgeand intelligence, which may be applied to the solution of world problems inindustry, the environment and the community. It also focusses on theknowledge-transfer methodologies and innovation strategies employed to makethis happen effectively. The combination of intelligent systems tools and a broadrange of applications introduces a need for a synergy of disciplines from science,technology, business and the humanities. The series will include conferenceproceedings, edited collections, monographs, handbooks, reference books, andother relevant types of book in areas of science and technology where smartsystems and technologies can offer innovative solutions.

High quality content is an essential feature for all book proposals accepted for theseries. It is expected that editors of all accepted volumes will ensure thatcontributions are subjected to an appropriate level of reviewing process and adhereto KES quality principles.

More information about this series at http://www.springer.com/series/8767

Giuseppe De Pietro ⋅ Luigi GalloRobert J. Howlett ⋅ Lakhmi C. JainEditors

Intelligent InteractiveMultimedia Systemsand Services 2016

123

EditorsGiuseppe De PietroInstitute of High Performance Computingand Networking

National Research CouncilNaplesItaly

Luigi GalloInstitute of High Performance Computingand Networking

National Research CouncilNaplesItaly

Robert J. HowlettKES InternationalShoreham-by-seaUK

Lakhmi C. JainUniversity of CanberraCanberraAustralia

and

Bournemouth UniversityPooleUK

and

KES InternationalShoreham-by-seaUK

ISSN 2190-3018 ISSN 2190-3026 (electronic)Smart Innovation, Systems and TechnologiesISBN 978-3-319-39344-5 ISBN 978-3-319-39345-2 (eBook)DOI 10.1007/978-3-319-39345-2

Library of Congress Control Number: 2016940122

© Springer International Publishing Switzerland 2016This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or partof the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmissionor information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilarmethodology now known or hereafter developed.The use of general descriptive names, registered names, trademarks, service marks, etc. in thispublication does not imply, even in the absence of a specific statement, that such names are exempt fromthe relevant protective laws and regulations and therefore free for general use.The publisher, the authors and the editors are safe to assume that the advice and information in thisbook are believed to be true and accurate at the date of publication. Neither the publisher nor theauthors or the editors give a warranty, express or implied, with respect to the material contained herein orfor any errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by Springer NatureThe registered company is Springer International Publishing AG Switzerland

Preface

Dear Readers,We introduce to you a series of carefully selected papers presented during the

9th KES International Conference on Intelligent Interactive Multimedia Systemsand Services (IIMSS-16).

At a time when computers are more widespread than ever and computer usersrange from highly qualified scientists to non-computer expert professionals, Intel-ligent Interactive Systems are becoming a necessity in modern computer systems.The solution of “one-fits-all” is no longer applicable to wide ranges of users ofvarious backgrounds and needs. Therefore, one important goal of many intelligentinteractive systems is dynamic personalization and adaptivity to users. MultimediaSystems refer to the coordinated storage, processing, transmission, and retrieval ofmultiple forms of information, such as audio, image, video, animation, graphics,and text. The growth rate of multimedia services has become explosive, as tech-nological progress matches consumer needs for content.

The IIMSS-16 conference took place as part of the Smart Digital Futures 2016multi-theme conference, which groups AMSTA-16, IDT-16, InMed-16, andSEEL-16 with IIMSS-16 in one venue. It was a forum for researchers and scientiststo share work and experiences on intelligent interactive systems and on multimediasystems and services. It included a general track and nine invited sessions.

The general track (Chaps. “Analysis of Similarity Measurements in CBIR UsingClustered Tamura Features for Biomedical Images”–“How to Manage Keys andReconfiguration in WSNs Exploiting SRAM Based PUFs”) focused on intelligentimage or video storage, retrieval, transmission, and analysis. The invited session“Intelligent Video Processing and Transmission Systems” (Chaps. “Fast SalientObject Detection in Non-stationary Video Sequences Based on Spatial SaliencyMaps”–“Development Prospects of the Visual Data Compression Technologies andAdvantages of New Approaches”) specifically focused on functionalities and archi-tectures of systems for video processing and transmission. The invited session“Innovative Information Services for Advanced Knowledge Activity” (Chaps.“A Near-far Resistant Preambleless Blind Receiver with Eigenbeams Applicable to

v

Sensor Networks”–“Trends in Teaching/Learning Research Through Analysis ofConference Presentation Articles”) focused on novel functionalities for informationservices. The invited session “Autonomous System” (Chaps. “Motion Prediction forShip-Based Autonomous Air Vehicle Operations”–“Active Suspension InvestigationUsing Physical Networks”) considered issues such as motion prediction, operatingsystems, and networks for what concerns autonomous systems. The invited session“Mobility Data Analysis and Mining” (Chaps. “Automatic Generation of TrajectoryData Warehouse Schemas”–“A Survey on Web Service Mining Using QoS andRecommendation Based on Multidimensional Approach”) focused on novel mod-elling and analysis approaches for mobility data. The invited session “IntelligentComputer Systems Enhancing Creativity” (Chaps. “Mapping and PocketingTechniques for Laser Marking of 2D Shapes on 3D Curved Surfaces”–“Experience-driven Framework for Technologically-enhanced Environments: KeyChallenges and Potential Solutions”) provided insight into the most recent efforts,challenges, and best practices across the fields of computer-aided creativity andinnovation. The invited session “Internet of Things: Architecture, Technologies andApplications” (Chaps. “Touchless Disambiguation Techniques for WearableAugmented Reality Systems”–“Opinions Analysis in Social Networks for CulturalHeritage Applications”) focused on IoT approaches, especially considering culturalheritage scenarios. The invited session “Interactive Cognitive Systems” (Chaps.“A Forward-Selection Algorithm for SVM-Based Question Classification inCognitive Systems”–“A Model of a Social Chatbot”) focused on adaptive andhuman-like cognitive systems and on artificial intelligence systems and robotics. Theinvited session “Smart Environments and Information Systems” (Chaps. “AnExperience ofEngineering ofMAS for Smart Environments: ExtensionofASPECS”–“Soft Sensor Network For Environmental Monitoring”) discussed the requirementsof the information systems supporting smart environments, aswell as themethods andtechniques that are currently being explored. Finally, the invited session “New Tech-nologies and Virtual Reality in Health Systems” (Chaps. “The Use of Eye Tracking(ET) in Targeting Sports: A Review of the Studies on Quiet Eye (QE)”–“TheElapsedTimeDuringaVirtualRealityTreatment forStressfulProcedures.APoolAnalysis on Breast Cancer Patients During Chemotherapy”) focused on advancedfunctionalities for VR-based applications in health care.

Our gratitude goes to many people who have greatly contributed to puttingtogether a fine scientific programme and exciting social events for IIMSS 2016. Weacknowledge the commitment and hard work of the programme chairs and theinvited session organizers. They have kept the scientific programme in focus andmade the discussions interesting and valuable. We recognize the excellent job doneby the programme committee members and the extra reviewers. They evaluated allthe papers on a very tight time schedule. We are grateful for their dedication andcontributions. We could not have done it without them. More importantly, we thankthe authors for submitting and trusting their work to the IIMSS conference.

vi Preface

We hope that readers will find in this book an interesting source of knowledge infundamental and applied facets of intelligent interactive multimedia and, maybe,even some motivation for further research.

Giuseppe De PietroLuigi Gallo

Robert J. HowlettLakhmi C. Jain

Preface vii

Organization

Honorary Chairs

Toyohide Watanabe, Nagoya University, JapanLakhmi C. Jain, University of Canberra, Canberra, Australia;Bournemouth University, UK

Co-General Chairs

Giuseppe De Pietro, National Research Council, ItalyLuigi Gallo, National Research Council, Italy

Executive Chair

Robert J. Howlett, University of Bournemouth, UK

Programme Chair

Antonino Mazzeo, University of Naples Federico II, Italy

Publicity Chair

Giuseppe Caggianese, National Research Council, Italy

ix

Invited Session Chairs

Intelligent Video Processing and Transmission Systems

Margarita N. Favorskaya, Siberian State Aerospace University, RussiaLakhmi C. Jain, University of Canberra, Australia and Bournemouth University,UKMikhail Sergeev, Saint-Petersburg State University of Aerospace Instrumentation,ITMO University, Russia

Innovative Information Services for Advanced Knowledge Activity

Koichi Asakura, Daido University, JapanToyohide Watanabe, Nagoya Industrial Research Institute, Japan

Autonomous System

Milan Simic, RMIT University, AustraliaReza Nakhaie Jazar, RMIT University, Australia

Mobility Data Analysis and Mining

Jalel Akaichi, King Khalid University, Saudi Arabia

Intelligent Computer Systems Enhancing Creativity

Raffaele De Amicis, Graphitech, ItalyDavid Oyarzun, Vicomtech-IK4, Spain

Internet of Things: Architecture, Technologies and Applications

Francesco Piccialli, University of Naples Federico II, ItalyAngelo Chianese, University of Naples Federico II, Italy

Interactive Cognitive Systems

Ignazio Infantino, National Research Council, ItalyMassimo Esposito, National Research Council, Italy

Smart Environments and Information Systems

Massimo Cossentino, National Research Council of Italy, ItalyVincent Hilaire, Université de Belfort-Montbeliard, FranceJuan Pavon, Universidad Complutense de Madrid, Spain

New Technologies and Virtual Reality in Health Systems

Antonio Giordano, College of Science and Technology, Temple University, USA

x Organization

International Programme Committee

Jalel Akaichi, King Khalid University, Saudi ArabiaRaffaele De Amicis, Graphitech, ItalyKoichi Asakura, Daido University, JapanMonica Bianchini, Università degli Studi di Siena, ItalyEl Fazziki Aziz, Cadi Ayyad University of Marrakesh, MoroccoAngelo Chianese, University of Naples Federico II, ItalyMassimo Cossentino, National Research Council of Italy, ItalyMario Döller, University of applied science Kufstein Tirol, AustriaDinu Dragan, Faculty of Technical Sciences, University of Novi Sad, SerbiaMassimo Esposito, National Research Council of Italy, ItalyMargarita Favorskaya, Siberian State Aerospace University, RussiaColette Faucher, LSIS-Polytech Marseille, FranceChristos Grecos, Sohar University, OmanVincent Hilaire, Université de Belfort-Montbeliard, FranceKatsuhiro Honda, Osaka Prefecture University, JapanHsiang-Cheh Huang, National University of Kaohsiung, TaiwanIgnazio Infantino, National Research Council of Italy, ItalyLakhmi C. Jain, University of Canberra, Australia and Bournemouth University,UKReza Nakhaie Jazar, SAMME, RMIT University, AustraliaDimitris Kanellopoulos, University of Patras, GreeceChengjun Liu, New Jersey Institute of Technology, USACristian Mihăescu, University of Craiova, RomaniaVincent Oria, New Jersey Institute of Technology, USADavid Oyarzun, Vicomtech-IK4, SpainJuan Pavon, Universidad Complutense de Madrid, SpainFrancesco Piccialli, University of Naples Federico II, ItalyRadu-Emil Precup, Politehnica University of Timisoara, RomaniaLuca Sabatucci, National Research Council of Italy, ItalyMohammed Sadgal, Cadi Ayyad University of Marrakesh, MoroccoA. Sadiq, Cadi Ayyad University of Marrakesh, MoroccoMikhail Sergeev, ITMO University, RussiaMilan Simic, SAMME, RMIT University, Melbourne, AustraliaTaketoshi Ushiama, Kyushu University, JapanToyohide Watanabe, Nagoya Industrial Science Research Institute, Japan

Organization xi

Contents

Analysis of Similarity Measurements in CBIR Using ClusteredTamura Features for Biomedical Images . . . . . . . . . . . . . . . . . . . . . . . 1Nadia Brancati and Francesco Camastra

2-Stripes Block-Circulant LDPC Codes for Single BurstsCorrection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Evgenii Krouk and Andrei Ovchinnikov

Data Dictionary Extraction for Robust Emergency Detection . . . . . . . . 25Emanuele Cipolla and Filippo Vella

SmartCARE—An ICT Platform in the Domain of StrokePathology to Manage Rehabilitation Treatmentand Telemonitoring at Home . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39Francesco Adinolfi, Giuseppe Caggianese, Luigi Gallo, Juan Grosso,Francesco Infarinato, Nazzareno Marchese, Patrizio Saleand Emiliano Spaltro

Optimal Design of IPsec-Based Mobile Virtual PrivateNetworks for Secure Transfer of Multimedia Data . . . . . . . . . . . . . . . . 51Alexander V. Uskov, Natalia A. Serdyukova, Vladimir I. Serdyukov,Adam Byerly and Colleen Heinemann

Malicious Event Detecting in Twitter Communities . . . . . . . . . . . . . . . 63Flora Amato, Giovanni Cozzolino, Antonino Mazzeoand Sara Romano

Adopting Decision Tree Based Policy Enforcement Mechanismto Protect Reconfigurable Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73Mario Barbareschi, Antonino Mazzeo and Salvatore Miranda

Arabic Named Entity Recognition—A Survey and Analysis . . . . . . . . . 83Amal Dandashi, Jihad Al Jaam and Sebti Foufou

xiii

Exploitation of Web Resources Towards Increased Conversionsand Effectiveness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97Jarosław Jankowski, Jarosław Wątróbski, Paweł Ziembaand Wojciech Sałabun

How to Manage Keys and Reconfiguration in WSNs ExploitingSRAM Based PUFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109Domenico Amelino, Mario Barbareschi, Ermanno Battistaand Antonino Mazzeo

Fast Salient Object Detection in Non-stationary Video SequencesBased on Spatial Saliency Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121Margarita Favorskaya and Vladimir Buryachenko

Global Motion Estimation Using Saliency Maps in Non-stationaryVideos with Static Scenes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133Margarita Favorskaya, Vladimir Buryachenko and Anastasia Tomilina

SVM-Based Cancer Grading from Histopathological ImagesUsing Morphological and Topological Features of Glandsand Nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145Catalin Stoean, Ruxandra Stoean, Adrian Sandita, Daniela Ciobanu,Cristian Mesina and Corina Lavinia Gruia

On Preservation of Video Data When Transmitting in Systemsthat Use Open Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157Anton Vostrikov, Mikhail Sergeev and Nikolaj Solovjov

An Invariant Subcode of Linear Code . . . . . . . . . . . . . . . . . . . . . . . . . 169Sergei V. Fedorenko and Eugenii Krouk

Development Prospects of the Visual Data CompressionTechnologies and Advantages of New Approaches . . . . . . . . . . . . . . . . 179Anton Vostrikov and Mikhail Sergeev

A Near-Far Resistant Preambleless Blind Receiverwith Eigenbeams Applicable to Sensor Networks . . . . . . . . . . . . . . . . . 191Kuniaki Yano and Yukihiro Kamiya

A New Approach for Subsurface Wireless Sensor Networks . . . . . . . . . 201Hikaru Koike and Yukihiro Kamiya

Implementation of Mobile Sensing Platform with a Tree BasedSensor Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213Katsuhiro Naito, Shunsuke Tani and Daichi Takai

Prototype Implementation of Actuator Sensor Networkfor Agricultural Usages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227Takuya Wada and Katsuhiro Naito

xiv Contents

Development of Multi-hop Field Sensor Networkswith Arduino Board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241Tomoya Ogawa and Katsuhiro Naito

Communication Simulator with Network Behavior LoggingFunction for Supporting Network Construction Exercisefor Beginners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253Yuichiro Tateiwa and Naohisa Takahashi

A New Method to Apply BRAKE to Sensor NetworksAiming at Power Saving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267Minori Kinose and Yukihiro Kamiya

A Tool for Visualization of Meteorological Data Studiedfor Integration in a Multi Risk Management System . . . . . . . . . . . . . . 275Emanuele Cipolla, Riccardo Rizzo, Dario Stabile and Filippo Vella

An Algorithm for Calculating Simple Evacuation Routesin Evacuation Guidance Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287Koichi Asakura and Toyohide Watanabe

SkyCube-Tree Based Group-by Query Processing in OLAPSkyline Cubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297Hideki Sato and Takayuki Usami

Trends in Teaching/Learning Research Through Analysisof Conference Presentation Articles . . . . . . . . . . . . . . . . . . . . . . . . . . . 309Toyohide Watanabe

Motion Prediction for Ship-Based Autonomous Air VehicleOperations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323Ameer A. Khan, Kaye E. Marion, Cees Bil and Milan Simic

The Effect of Receding Horizon Pure Pursuit Controlon Passenger Comfort in Autonomous Vehicles . . . . . . . . . . . . . . . . . . 335Mohamed Elbanhawi, Milan Simic and Reza Jazar

From Automotive to Autonomous: Time-TriggeredOperating Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347Maria Spichkova, Milan Simic and Heinz Schmidt

Active Suspension Investigation Using Physical Networks . . . . . . . . . . . 361Milan Simic

Automatic Generation of Trajectory Data Warehouse Schemas. . . . . . . 373Nouha Arfaoui and Jalel Akaichi

A Taxonomy of Support Vector Machine for Event StreamsClassification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385Hanen Bouali, Yasser Al Mashhour and Jalel Akaichi

Contents xv

Social Networks Security Policies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395Zeineb Dhouioui, Abdullah Ali Alqahtani and Jalel Akaichi

Recommending Multidimensional Spatial OLAP Queries . . . . . . . . . . . 405Olfa Layouni, Fahad Alahmari and Jalel Akaichi

Modeling Moving Regions: Colorectal Cancer Case Study . . . . . . . . . . 417Marwa Massaâbi and Jalel Akaichi

A UML/MARTE Extension for Designing Energy Harvestingin Wireless Sensor Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427Raoudha Saida, Yessine Hadj Kacem, M.S. BenSalehand Mohamed Abid

A Survey on Web Service Mining Using QoSand Recommendation Based on Multidimensional Approach . . . . . . . . 439Ilhem Feddaoui, Faîçal Felhi, Imran Hassan Bergi and Jalel Akaichi

Mapping and Pocketing Techniques for Laser Markingof 2D Shapes on 3D Curved Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . 451Federico Devigili, Davide Lotto and Raffaele de Amicis

3DHOG for Geometric Similarity Measurement and Retrievalon Digital Cultural Heritage Archives . . . . . . . . . . . . . . . . . . . . . . . . . 459Reimar Tausch, Hendrik Schmedt, Pedro Santos, Martin Schröttnerand Dieter W. Fellner

Computer Aided Process as 3D Data Provider in FootwearIndustry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471Bita Ture Savadkoohi and Raffaele De Amicis

CoolTour: VR and AR Authoring Tool to Create CulturalExperiences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483Nagore Barrena, Andrés Navarro, Sara García and David Oyarzun

Emotional Platform for Marketing Research . . . . . . . . . . . . . . . . . . . . 491Andres Navarro, Catherine Delevoye and David Oyarzun

Gamification as a Key Enabling Technology for Image Sensingand Content Tagging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503Bruno Simões and Raffaele De Amicis

Bridging Heritage and Tourist UX: A Socially-Driven Perspective . . . . 515Paola La Scala, Bruno Simões and Raffaele De Amicis

A Personalised Recommender System for Tourists on City Trips:Concepts and Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525Petr Aksenov, Astrid Kemperman and Theo Arentze

xvi Contents

Experience-Driven Framework for Technologically-EnhancedEnvironments: Key Challenges and Potential Solutions. . . . . . . . . . . . . 537Bruno Simões and Raffaele De Amicis

Touchless Disambiguation Techniques for Wearable AugmentedReality Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547Giuseppe Caggianese, Luigi Gallo and Pietro Neroni

System Architecture and Functions of Internet-Based Productionof Tailor-Made Clothes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 557Petra Perner

Influence of Some Parameters on Visiting Style Classificationin a Cultural Heritage Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567Salvatore Cuomo, Pasquale De Michele, Ardelio Gallettiand Giovanni Ponti

Opinions Analysis in Social Networks for Cultural HeritageApplications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577Flora Amato, Giovanni Cozzolino, Sergio Di Martino,Antonino Mazzeo, Vincenzo Moscato, Antonio Picariello,Sara Romano and Giancarlo Sperlí

A Forward-Selection Algorithm for SVM-Based QuestionClassification in Cognitive Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 587Marco Pota, Massimo Esposito and Giuseppe De Pietro

Supporting Autonomy in Agent Oriented Methodologies. . . . . . . . . . . . 599Valeria Seidita and Massimo Cossentino

A Data-Driven Approach to Dynamically Learn FocusedLexicons for Recognizing Emotions in Social Network Streams . . . . . . . 609Diego Frias and Giovanni Pilato

Disaster Prevention Virtual Advisors Through SoftSensor Paradigm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 619Agnese Augello, Umberto Maniscalco, Giovanni Pilatoand Filippo Vella

A Personal Intelligent Coach for Smart Embodied LearningEnvironments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 629Agnese Augello, Ignazio Infantino, Adriano Manfré,Giovanni Pilato, Filippo Vella, Manuel Gentile, Giuseppe Città,Giulia Crifaci, Rossella Raso and Mario Allegra

A Model of a Social Chatbot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 637Agnese Augello, Manuel Gentile, Lucas Weideveldand Frank Dignum

Contents xvii

An Experience of Engineering of MAS for Smart Environments:Extension of ASPECS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 649Philippe Descamps, Vincent Hilaire, Olivier Lamotteand Sebastian Rodriguez

A Norm-Based Approach for Personalising Smart Environments . . . . . 659Patrizia Ribino, Carmelo Lodato, Antonella Cavaleriand Massimo Cossentino

Adopting a Middleware for Self-adaptation in the Developmentof a Smart Travel System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 671L. Sabatucci, A. Cavaleri and M. Cossentino

Pentas: Using Satellites for Smart Sensing . . . . . . . . . . . . . . . . . . . . . . 683Lorena Otero-Cerdeira, Alma Gómez-Rodríguez,Francisco J. Rodríguez-Martínez, Juan Carlos González-Morenoand Arno Formella

A Multiple Data Stream Management Framework for AmbientAssisted Living Emulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695Jorge J. Gómez-Sanz and Pablo Campillo Sánchez

Soft Sensor Network for Environmental Monitoring . . . . . . . . . . . . . . . 705Umberto Maniscalco, Giovanni Pilato and Filippo Vella

The Use of Eye Tracking (ET) in Targeting Sports: A Reviewof the Studies on Quiet Eye (QE). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715Dario Fegatelli, Francesco Giancamilli, Luca Mallia, Andrea Chiricoand Fabio Lucidi

The Elapsed Time During a Virtual Reality Treatmentfor Stressful Procedures. A Pool Analysis on Breast CancerPatients During Chemotherapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 731A. Chirico, M. D’Aiuto, M. Pinto, C. Milanese, A. Napoli, F. Avino,G. Iodice, G. Russo, M. De Laurentiis, G. Ciliberto, A. Giordanoand F. Lucidi

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 739

xviii Contents

Analysis of Similarity Measurementsin CBIR Using Clustered Tamura Featuresfor Biomedical Images

Nadia Brancati and Francesco Camastra

Abstract Content based image retrieval (CBIR) is an important research topic in

many applications, in particular in the biomedical field. In this domain, the CBIR has

the aim of helping to improve the diagnosis, retrieving images of patients for which

a diagnosis has already been made, similar to the current image. The main issue of

CBIR is the selection of the visual contents (feature descriptors) of the images to be

extracted for a correct image retrieval. The second issue is the choice of the similarity

measurement to use to compare the feature descriptors of the query image to ones

of the other images of the database. This paper focuses on a comparison among

different similarity measurements in CBIR, with particular interest to a biomedical

images database. The adopted technique for CBIR is based on clustered Tamura

features. The selected similarity measurements are used both to evaluate the adopted

technique for CBIR and to estimate the stability of the results. A comparison with

some methods in literature has been carried out, showing the best results for the

proposed technique.

Keywords Content based image retrieval ⋅ Tamura features ⋅ Similarity

measurement ⋅ CBIR by clustering

1 Introduction and Background

In the biomedical field, information systems help to improve the efficiency and the

quality of a diagnosis. In particular, for the clinical decision-making process it can be

very useful to find images with characteristics similar (same anatomic region, same

N. Brancati (✉)

Institute for High Performance Computing and Networking,

National Research Council of Italy (ICAR-CNR), Naples, Italy

e-mail: nadia.brancati@cnr.it; nadia.brancati@nai.icar.cnr.it

F. Camastra

Department of Science and Technology,

University of Naples Parthenope, Naples, Italy

e-mail: camastra@ieee.org

© Springer International Publishing Switzerland 2016

G. De Pietro et al. (eds.), Intelligent Interactive Multimedia Systemsand Services 2016, Smart Innovation, Systems and Technologies 55,

DOI 10.1007/978-3-319-39345-2_1

1

2 N. Brancati and F. Camastra

disease, ...) to a given image. For this purpose a content based image retrieval (CBIR)

system could be used [11, 14, 19], both to benefit the management of increasingly

large image collections, and to support clinical care, biomedical research, and edu-

cation. However, although the number of experimental algorithms comprehending

specific problems and databases is growing, few systems exist with relative success

[3, 7, 29]. So, biomedical applications are one of the priority areas where CBIR

can meet more success outside the experimental sphere, due to population aging in

developed countries.

The CBIR is the technique that allows retrieving images similar to a query image,

in a large unannotated database. The retrieval of the similar images is based on the

extraction of some visual contents of the images, called feature descriptors.

Many different feature descriptors have been proposed and used in the past years

[19, 25, 26]. These feature descriptors are usually low level features, easy to extract

and they are mainly of two types: global as shape, color or texture [1, 5, 6, 12, 24]

and local, that focus mainly on key points or salient patches [10, 15, 21, 22, 28].

After the choice of the most appropriate feature descriptors, these are extracted

both from all the images in the database and from the query image. At this point,

a similarity measurement should be chosen to compute the distance between the

feature descriptors of the query image and the feature descriptors of all the images in

the database. The choice of the appropriate similarity measurement could be another

crucial element for the correct design of a CBIR system [16].

In some cases, it results necessary to make CBIR techniques more efficient and

accurate, above all when the databases are very large. In this case, the aim is to

decrease the number of the images for which to compute the distance from the query

image. Many clustering techniques with some feature descriptors of the images can

be used [2, 8, 9, 18].

In this paper, the retrieval of images more similar to a query image is performed

using textural features, in particular Tamura features, that correspond to human

visual perception [27]. Then, a clustering using K-Means Algorithm [13] is carried

out to obtain homogeneous groups, based on Tamura features [23].

The main contribution of the paper is the validation of the use of Tamura features,

for content based image retrieval, in particular for biomedical databases, compared

with the use of local descriptors. Moreover, a comparison among different similarity

measurements is performed both to evaluate the adopted technique for CBIR and to

estimate the stability of the results.

2 CBIR Steps

CBIR technique proposed in this paper is composed by the following steps:

∙ the extraction of Tamura features from all the images in the database and from the

query image;

Analysis of Similarity Measurements in CBIR Using . . . 3

∙ the clustering of the Tamura features, extracted from all the images in the database,

using K-Means algorithm;

∙ the computation of five distance metrics both to evaluate the adopted technique

for CBIR and to estimate the stability of the results.

2.1 Tamura Features

Tamura features correspond to human visual perception. They were designed in

accord to psychological studies on the human perception and they capture the high-

level perceptual attributes of a texture. They define six textural features: coarseness,

directionality, contrast, roughness, line-likeness and regularity. The first three fea-

tures are the most similar to human visual perception, and they are considered in the

present work; they are extracted both from all images in the database and from the

query image. More details of these features can be found in [27].

Coarseness The aim of this feature is to find a repetitive pattern in the texture,

which can have several orders of magnitude, depending on whether you are in front

of a coarse or fine texture. So, operators to several orders of magnitude are computed.

If the texture is fine, the highest response will be given by the operator of magnitude

lower, vice versa, if the texture is coarse, the highest response will be given by the

operator of magnitude greater.

The computation of the coarseness is given from:

Fcrs =1

m × n

m∑

i=0

n∑

j=0Sbest(i, j)

where m × n is the resolution of the image and Sbest(i, j) is computed for each pixel

and it provides the information about the magnitude of the pattern. It is important to

underline that the coarseness feature is influenced both by the size of the pattern to

find and by its repetitiveness.

Contrast The contrast of Tamura features takes into account both the variation

range of the gray levels and the polarization of white and black pixels. A measure-

ment for the variation range is the variance 𝜎

2of the pixels of the image. In fact,

they measure the dispersion present in the distribution of the gray levels. However

this single measurement does not appear to be very significant when the image his-

togram shows a prominent peak towards white or towards black. A measurement

for the polarization of white and black pixels is given by the kurtosis, defined as

𝛼4 = 𝜇4∕𝜎4, where 𝜇4 is the moment of fourth order. At this point, the two measure-

ments are combined, obtaining the feature of the contrast:

Fcon =𝜎

(𝛼4)n

4 N. Brancati and F. Camastra

where n can be equal to 8, 4, 2, 1, 1∕2, 1∕4, 1∕8. In this paper, for the experiments,

n is set to 1.

Directionality The directionality is a feature that can be calculated from the

analysis of the Fourier spectrum. However, the features obtained by the Fourier spec-

trum do not behave in the same way as those calculable in the spatial domain. Tamura

preferred to get a global feature of the image, analysing the histogram of the direc-

tions of the edges, the form of the gradient image and the peaks of the histogram. In

particular, the sum of the moments of second order around each peak, from to a val-

ley to another valley is computed, and this measurement is defined in the following

way:

Fdir = 1 − rnpnp∑

p

∑

𝜙∈wp

(𝜙 − 𝜙p)2HD(𝜙)

where np is the number of the peaks, 𝜙p is the pth peak of HD, wp is the range of the

pth peak between two valleys, 𝜙 is the quantized directionality, r is a normalization

factor, related to the quantized levels of 𝜙.

2.2 Clustering

After the extraction of the Tamura features from all images in the database, these

features are clustered, using K-Means algorithm. For the experiments in this paper,

the number of clusters is set to 2. These information are saved in a data structure in

order to use them for the following experiments.

2.3 Distance Metrics

The choice of the similarity measurement is the second issue in CBIR. For the pro-

posed technique, first the distances between the feature descriptors of the query

image and the centroids of the clusters are computed. The cluster at minimum dis-

tance is selected. Then, the distances between the feature descriptors of the query

image and the feature descriptors of all the images of the selected cluster are com-

puted. The images with feature descriptors with small distance from the feature

descriptors of the query image are considered as the images more similar to the query

image. In this work, some distance metrics are used as similarity measurements. In

particular, well-known distance metrics are used [4]:

∙ Euclidean distance;

∙ City block distance;

∙ Minkowski distance, with order p = 3.

Analysis of Similarity Measurements in CBIR Using . . . 5

Moreover, let the vector of the feature descriptors of the query image be represented

by Q, and the vector of the feature descriptors of an image of the database be repre-

sented by I, two additional distance metrics are calculated:

Canberra distance, that normalizes each feature pair difference by dividing it by

the sum of a pair of feature descriptors:

D =n∑

i=1

(|Qi − Ii|)|Qi| + |Ii|

d1 distance, where the distance between two vectors of feature descriptors is

calculated based on the formula described in [21]:

D =n∑

i=1

|Qi − Ii||1 + Qi + Ii|

3 Experimental Results and Discussion

In order to analyse the performance of the proposed technique, some experiments

have been performed. The experiments have been conducted on the Open Access

Series of Imaging Studies (OASIS) [17]. It is a series of magnetic resonance imaging

(MRI) database that is publicly available for study and analysis. This dataset consists

of a cross-sectional collection of 421 subjects aged between 18 to 96 years, including

individuals with early-stage Alzheimer’s Disease (AD). For image retrieval purpose,

these 421 images are grouped into four categories (124, 102, 89, and 106 images)

based on the ventricular shape in the images. Sample images for each category are

displayed in the Fig. 1.

For the proposed CBIR technique, each image of the database is used as a query

image and the distances from the clusters are calculated, based on the distance met-

rics of the Sect. 2.3. When the cluster containing images more similar to the query

Fig. 1 Sample images from OASIS database (one image per category)

6 N. Brancati and F. Camastra

Fig. 2 The results for a query image of the group 1

Fig. 3 The results for a query image of the group 4

image is found, the distances between the query image and each image of the cluster

are computed, and the images more similar to the query are displayed.

Some results for two query image examples, for groups 1 and 4, are shown in the

Figs. 2 and 3.

The average retrieval precision (ARP) and the average retrieval rate (ARR) are

calculated, to evaluate the performance:

ARP = 1N

N∑

i=1PR(Qi)

||||m

Analysis of Similarity Measurements in CBIR Using . . . 7

ARR = 1N

N∑

i=1RE(Qi)

||||p

where Q is the query image and N is the total number of images in the database and

where precision (PR) and recall (RE) are:

PR(Q) =Number of Relevant Images RetrievedTotal Number of Images Retrieved

RE(Q) =Number of Relevant Images Retrieved

Total Number of Relevant Images in the Database

As specified in [20], to calculate precision and recall the number of images

retrieved should be specified, (e.g. precision with m = 20 images and recall with

p = 100 images are retrieved). So, for the current experiment, the number m of

images retrieved for precision is 100 and for recall is specific for each group, i.e.

p = 124 for the group 1, p = 102 for the group 2, p = 89 for the group 3, p = 106for the group 4.

In Tables 1 and 2 the results of ARP and ARR for all similarity measurements and

for all groups are reported. The performance of the proposed technique is better for

group 1 and 4, independently from the chosen similarity measurement. Moreover, all

similarity measurements for each group, provide values very similar among them,

showing a stability and a robustness of the technique. However, ARP and ARR show

a consistent behaviour for all groups: for the groups 1 and 2, they show the best

values for the City Block distance, for the group 3 the best values are given by the

Minkowski distance, for the group 4 are given by the d1 distance and for all groups

the best values of ARP are given by the City Block and Minkowski distance and the

best values of ARR are given by the City Block, Minkowski and Euclidean distance.

Finally, a comparison with some methods in literature are reported in Table 3

[28]. In order to compare with these methods, the number of retrieved images is 10(m = 10), as in [28]. As similarity measurement the Euclidean distance has been

chosen.

Table 1 Average retrieval precision (ARP) of all similarity measurements for each category

ARP (%)

Group 1 Group 2 Group 3 Group 4 TotalEuclidean 0.500 0.378 0.334 0.710 0.480

City Block 0.505 0.381 0.329 0.711 0.482Canberra 0.461 0.375 0.276 0.735 0.462

Minkowski 0.497 0.379 0.338 0.713 0.482d1 0.465 0.377 0.279 0.736 0.465

8 N. Brancati and F. Camastra

Table 2 Average retrieval rate (ARR) of all similarity measurements for each category

ARR (%)

Group 1 Group 2 Group 3 Group 4 Total

Euclidean 0.464 0.392 0.357 0.817 0.507City Block 0.465 0.394 0.351 0.817 0.507Canberra 0.425 0.388 0.297 0.843 0.488

Minkowski 0.463 0.388 0.358 0.819 0.507d1 0.428 0.391 0.301 0.844 0.491

Table 3 Comparison of the CBIR proposed technique with other methods in literature

ARP (m = 10) (%)

Group 1 Group 2 Group 3 Group 4 Total

CSLBP 0.46 0.36 0.29 0.4 0.38

LEPINV 0.48 0.34 0.29 0.41 0.38

LEPSEG 0.51 0.34 0.29 0.43 0.39

LBP 0.56 0.34 0.34 0.45 0.42

LMEBP 0.55 0.35 0.39 0.54 0.46

DLEP 0.51 0.37 0.38 0.53 0.45

CSLBCoP 0.55 0.39 0.38 0.64 0.49

CBIR

proposed

0.58 0.34 0.37 0.78 0.52

These methods are all based on local information of the pixels [28]:

∙ CSLBP: center simmetric local binary pattern;

∙ LEPINV: local edge pattern for image retrieval;

∙ LEPSEG: local edge pattern for segmentation;

∙ LBP: local binary pattern;

∙ LMEBP: local maximum edge binary pattern;

∙ DLEP: directional local extrema pattern;

∙ CSLBcoP: CSLBP + gray level co-occurrence matrix (GCLM).

The results show that the proposed CBIR technique outperforms the other meth-

ods for groups 1 and 4 and in terms of total ARP.

The promising results prove that the Tamura features represent good global visual

descriptors and that the local visual descriptors are less representative for the kind of

examined images. In particular, the difference between ventricular shape of healthy

subjects and subjects with early-stage Alzheimer’s Disease is well highlighted by

Tamura features, above all for the groups 1 and 4, for which the best results are

obtained.

Analysis of Similarity Measurements in CBIR Using . . . 9

4 Conclusion

In this paper a CBIR technique based on the clustering of the Tamura features has

been proposed. A biomedical database, containing MRI images (OASIS) has been

used for the experiments and different similarity measurements have been used both

to evaluate the proposed technique and to verify the stability of the technique. The

results show that the proposed technique is stable and robust, independently from the

selected distance metric; in fact both the Average Retrieval Precision (ARP) and the

Average Retrieval Rate (ARR) show similar results for all distance metrics. More-

over, the comparison with other methods in literature, that use local information of

the pixels, show that the proposed technique, based on global information, outper-

forms these methods in terms of ARP, with a number of retrieved images equal to 10.

References

1. Baraldi, A., Parmiggiani, F.: An investigation of the textural characteristics associated with

gray level cooccurrence matrix statistical parameters. IEEE Trans. Geosci. Remote Sensing

33(2), 293–304 (1995)

2. Chen, Y., Wang, J.Z., Krovetz, R.: Content-based image retrieval by clustering. In: Proceedings

of the 5th ACM SIGMM international workshop on Multimedia information retrieval. pp. 193–

200. ACM (2003)

3. Deserno, T.M., Güld, M.O., Plodowski, B., Spitzer, K., Wein, B.B., Schubert, H., Ney, H.,

Seidl, T.: Extended query refinement for medical image retrieval. J. Digital Imaging 21(3),

280–289 (2008)

4. Deza, M.M., Deza, E.: Encyclopedia of Distances. Springer (2009)

5. Ergen, B., Baykara, M.: Texture based feature extraction methods for content based medical

image retrieval systems. Bio-Med. Mater. Eng. 24(6), 3055–3062 (2014)

6. Haridas, K., Thanamani, A.S.: Well-organized content based image retrieval system in RGB

Color Histogram, Tamura Texture and Gabor Feature. Int. J. Adv. Res. Comput. Commun.

Engineering 3(10) (2014)

7. Hsu, W., Long, L.R., Antani, S., et al.: SPIRS: a framework for content-based image retrieval

from large biomedical databases. MedInfo 12, 188–192 (2007)

8. Jain, M., Singh, S.: A survey on: content based image retrieval systems using clustering tech-

niques for large data sets. Int. J. Managing Inf. Technol. (IJMIT) 3(4), 23–39 (2011)

9. Jain, M., Singh, S.: An experimental study on content based image retrieval based on number

of clusters using hierarchical clustering algorithm. Int. J. Sig. Process. Image Process. Pattern

Recogn. 7(4) (2014)

10. Jasmine, K.P., Kumar, P.R.: Multi-resolution joint LBP Histograms for biomedical image

retrieval. Int. J. Comput. Appl. 95(3) (2014)

11. Lehmann, T.M., Gold, M., Thies, C., Fischer, B., Spitzer, K., Keysers, D., Ney, H., Kohnen,

M., Schubert, H., Wein, B.B.: Content-based image retrieval in medical applications. Methods

Inf. Med. 43(4), 354–361 (2004)

12. Liu, G.H., Yang, J.Y.: Content-based image retrieval using color difference histogram. Pattern

Recogn. 46(1), 188–198 (2013)

13. Lloyd, S.P.: Least squares quantization in PCM. IEEE Trans. Inf. Theory 28(2), 129–137

(1982)

14. Long, L.R., Antani, S., Deserno, T.M., Thoma, G.R.: Content-based image retrieval in medi-

cine: retrospective assessment, state of the art, and future directions. Int. J. Healthc. Inf. Syst.

Inf. Official Publ. Inf. Resour. Manag. Assoc. 4(1), 1 (2009)

10 N. Brancati and F. Camastra

15. Lowe, D.G.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis.

60(2), 91–110 (2004)

16. Malik, F., Baharudin, B.: Analysis of distance metrics in content-based image retrieval using

statistical quantized histogram texture features in the DCT domain. J. King Saud Univ.-

Comput. Inf. Sci. 25(2), 207–218 (2013)

17. Marcus, D.S., Wang, T.H., Parker, J., Csernansky, J.G., Morris, J.C., Buckner, R.L.: Open

access series of imaging studies (OASIS): cross-sectional MRI data in young, middle aged,

nondemented, and demented older adults. J. Cogn. Neurosci. 19(9), 1498–1507 (2007)

18. Mishra, P., Sonam, M., Vijayalakshmi, M.S.: Content based image retrieval using clustering

technique: a survey. Int. J. Res. Comput. Eng. Electron. 3(2) (2014)

19. Müller, H., Michoux, N., Bandon, D., Geissbuhler, A.: A review of content-based image

retrieval systems in medical applications—clinical benefits and future directions. Int. J. Med.

Inf. 73(1), 1–23 (2004)

20. Müller, H., Müller, W., Squire, D.M., Marchand-Maillet, S., Pun, T.: Performance evaluation in

content-based image retrieval: overview and proposals. Pattern Recogn. Lett. 22(5), 593–601

(2001)

21. Murala, S., Maheshwari, R., Balasubramanian, R.: Directional binary wavelet patterns for bio-

medical image indexing and retrieval. J. Med. Syst. 36(5), 2865–2879 (2012)

22. Murala, S., Wu, Q.: Local mesh patterns versus local binary patterns: biomedical image index-

ing and retrieval. IEEE J. Biomed. Health Inf. 18(3), 929–938 (2014)

23. Murthy, V., Vamsidhar, E., Kumar, J.S., Rao, P.S.: Content based image retrieval using hierar-

chical and k-means clustering techniques. Int. J. Eng. Sci. Technol. 2(3), 209–212 (2010)

24. Puviarasan, N., Bhavani, R., Vasanthi, A.: Image retrieval using combination of texture and

shape features. Image 3(3) (2014)

25. Rui, Y., Huang, T.S., Chang, S.F.: Image retrieval: current techniques, promising directions,

and open issues. J. Vis. Commun. Image Represent. 10(1), 39–62 (1999)

26. Smeulders, A.W., Worring, M., Santini, S., Gupta, A., Jain, R.: Content-based image retrieval

at the end of the early years. IEEE Trans. Pattern Anal. Mach. Intell. 22(12), 1349–1380 (2000)

27. Tamura, H., Mori, S., Yamawaki, T.: Textural features corresponding to visual perception.

IEEE Trans. Syst. Man Cybern. 8(6), 460–473 (1978)

28. Verma, M., Raman, B.: Center symmetric local binary co-occurrence pattern for texture, face

and bio-medical image retrieval. J. Vis. Commun. Image Represent. 32, 224–236 (2015)

29. Xue, Z., Long, L.R., Antani, S., Jeronimo, J., Thoma, G.R.: A web-accessible content-based

cervicographic image retrieval system. In: Medical imaging. International Society for Optics

and Photonics, pp. 691907–691907 (2008)

2-Stripes Block-Circulant LDPC Codesfor Single Bursts Correction

Evgenii Krouk and Andrei Ovchinnikov

Abstract In this paper the low-density parity-check (LDPC) codes are considered

applied to correction of error bursts. Errors grouping and forming of so-called bursts

are typical effect in real communication and data storage systems, however, this effect

is typically ignored, and the coding task is reduced to correction of independent

errors, which makes the practical characteristics of coding systems worse compar-

ing to possibly reachable. Nevertheless, LDPC codes are able to protect from burst

errors as well as independent ones. The main result of the paper is dedicated to eval-

uation of maximum correctable burst length of Gilbert codes, which are the 2-stripes

special case of LDPC block-permutation codes, the construction which is often used

in modern practical applications and research.

Keywords LDPC codes ⋅ Bursts-correcting codes ⋅ Gilbert codes

1 Introduction

During the development of modern practical communication systems the channel

models which are commonly used (binary-symmetric channel or Gaussian channel)

are often inadequate since they consider independent errors. At the same time in

real communication channels the effect of “memory” occurs (for example due to

fading [1]) leading to the dependencies between erroneous symbols. To fight with

such errors grouping the interleaving procedure is often used [1].

However, usage of interleaver leads to typical channel behaviour loss, the chan-

nel is transformed to memoryless, this decreases the possible transmission rates,

and increases the complexity and delay of transmitter and receiver [2, 3]. This is

E. Krouk ⋅ A. Ovchinnikov (✉)

Saint-Petersburg State University of Aerospace Instrumentation,

B.Morskaya 67, 190000 Saint Petersburg, Russia

e-mail: mldoc@ieee.org

E. Krouk

e-mail: ekrouk@vu.spb.ru

© Springer International Publishing Switzerland 2016

G. De Pietro et al. (eds.), Intelligent Interactive Multimedia Systemsand Services 2016, Smart Innovation, Systems and Technologies 55,

DOI 10.1007/978-3-319-39345-2_2

11

12 E. Krouk and A. Ovchinnikov

because the classical coding theory usually proposes code constructions for inde-

pendent errors which are simpler to analyse. So the important task is to construct

coding schemes oriented on typical channel errors, in particular, on correcting the

error bursts, that is, the error patterns when first and last erroneous symbols are no

far than some value b from each other (and which is called the burst length). Besides,

the effect of errors grouping is typical for data storage systems.

In the coding theory the classes of burst-correcting codes are known. For exam-

ple, these are Fire codes or Reed-Solomon codes [4]. During the last decades a lot

attention was given to low-density parity-check (LDPC) codes, particularly block-

permutation constructions [5]. Gilbert codes which are considered in this paper

are the simple special case of such construction and were proposed initially for

burst-correction. However, the exact burst-correction capability of these codes was

unknown.

The paper is organized as follows. Section 2 describes Gilbert codes and known

estimations of its burst-correction capability. In Sect. 3 the procedure is derived

allowing computation of exact value of maximum correctable burst length. The

Sect. 4 concludes the paper.

2 Gilbert Codes

LDPC-codes were invented by Gallager [6, 7] and later investigated in many works

[8–11]. While possessing comparatively poor minimal distance, these codes, how-

ever, provide high error-correction capability with very low decoding complexity. It

was shown that LDPC codes may overcome turbo-codes and approach to channel

capacity [12]. Additionally, some LDPC constructions (and block-permutation con-

structions as well) are cyclic or quasi-cyclic, allowing effective coder and decoder

implementation.

Block-permutation codes are one of the most prominent and widely used class

of LDPC codes [3, 5, 13]. The simple special case of this class are Gilbert codes,

which were proposed in [14] as burst-correction codes. Gilbert codes may be defined

by the parity-check matrix H𝓁 ,

H𝓁 =[

Im Im Im … ImIm C C2 … C𝓁−1

], (1)

where Im is (m × m)-unity matrix, C is (m × m)-matrix of cyclic permutation:

C =

⎡⎢⎢⎢⎢⎣

0 0 0 … 0 11 0 0 … 0 00 1 0 … 0 0… … … … … …0 0 0 … 1 0

⎤⎥⎥⎥⎥⎦

, (2)

and 𝓁 ≤ m.

2-Stripes Block-Circulant LDPC Codes for Single Bursts Correction 13

Many works were dedicated to estimation of burst-correcting capability of these

codes, as well as their modifications and extensions [15–19]. In [20] the estimation

of maximum correctable burst length b for the codes defined by matrix (1) is given

by inequality

b ≤ min𝛾∈{0,𝓁−2}

max{𝛾 − 1,m − 𝛾 − 1}. (3)

However, estimation (3) is not correct, giving only the lower bound of the maximum

correctable burst length. The exactness of this estimation decreases with growth of 𝓁.

In the next section we will give the method of exact evaluation of b.

3 Burst-Correction Capability of Gilbert Codes

The main result of this section and paper is the following theorem.

Theorem 1 Code with parity-check matrix H𝓁 defined by (1) can correct singlebursts of maximal length b𝓁 , where b𝓁 is calculated by the first satisfied condition:

1. b3 = m − 1, m is odd.2. If 𝓁 > ⌈m∕2⌉ + 1, then

{b𝓁 = m − ⌈m∕2⌉ + 1, m odd,b𝓁 = m∕2 − 1, m even.

3. If 𝓁 ≤ ⌈m∕2⌉ + 1, then

b𝓁 = m − 𝓁 + 1, if m ⋮ (𝓁 − 1),b𝓁 = m − 𝓁 + 1, if ∃k > 0 ∶ (m − 𝓁 + 3 − k ⋅ (𝓁 − 1)) ⋮ (𝓁 − 2),b𝓁 = m − 𝓁 + 2, if ∃k > 0 ∶ (m − k ⋅ (𝓁 − 3)) ⋮ (𝓁 − 2),b𝓁 = m − 𝓁 + 2, if ∃k > 0 ∶ (m − k ⋅ (𝓁 − 2)) ⋮ (𝓁 − 1),b𝓁 = m − 𝓁 + 2, if ∃k > 0 ∶ (m − k ⋅ (𝓁 − 1)) ⋮ (𝓁 − 2).

4. If all preceding conditions are unsatisfied, then:

b𝓁 = b𝓁−1.

Proof To prove the statement of the theorem we will introduce some notations and

prove some lemmas. Represent the matrix (1) as H𝓁 = [h0, h1,… , h𝓁−1], where h𝛾

—

(2m × m)-block-column, which we will call as block.

The code can correct single error burst of length b, if and only if all packets of

length b are in different cosets, i.e. there are no two error vectors e1 and e2 (forming

the bursts of length no more than b), such that

e1 ⋅ HT𝓁 = e2 ⋅ HT

𝓁 . (4)