Computational Intelligence for Pattern Recognition PDF
Preview Computational Intelligence for Pattern Recognition
Studies in Computational Intelligence 777 Witold Pedrycz · Shyi-Ming Chen Editors Computational Intelligence for Pattern Recognition Studies in Computational Intelligence Volume 777 Series editor Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland e-mail: [email protected] The series “Studies in Computational Intelligence” (SCI) publishes new develop- mentsandadvancesinthevariousareasofcomputationalintelligence—quicklyand with a high quality. The intent is to cover the theory, applications, and design methods of computational intelligence, as embedded in the fields of engineering, computer science, physics and life sciences, as well as the methodologies behind them. The series contains monographs, lecture notes and edited volumes in computational intelligence spanning the areas of neural networks, connectionist systems, genetic algorithms, evolutionary computation, artificial intelligence, cellular automata, self-organizing systems, soft computing, fuzzy systems, and hybrid intelligent systems. Of particular value to both the contributors and the readership are the short publication timeframe and the world-wide distribution, which enable both wide and rapid dissemination of research output. More information about this series at http://www.springer.com/series/7092 Witold Pedrycz Shyi-Ming Chen (cid:129) Editors Computational Intelligence for Pattern Recognition 123 Editors Witold Pedrycz Shyi-Ming Chen University of Alberta National Taiwan University of Science Edmonton, AB andTechnology Canada Taipei Taiwan ISSN 1860-949X ISSN 1860-9503 (electronic) Studies in Computational Intelligence ISBN978-3-319-89628-1 ISBN978-3-319-89629-8 (eBook) https://doi.org/10.1007/978-3-319-89629-8 LibraryofCongressControlNumber:2018937687 ©SpringerInternationalPublishingAG,partofSpringerNature2018 Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpart of 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 transmission orinformationstorageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilar methodologynowknownorhereafterdeveloped. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publicationdoesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfrom therelevantprotectivelawsandregulationsandthereforefreeforgeneraluse. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authorsortheeditorsgiveawarranty,expressorimplied,withrespecttothematerialcontainedhereinor for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictionalclaimsinpublishedmapsandinstitutionalaffiliations. Printedonacid-freepaper ThisSpringerimprintispublishedbytheregisteredcompanySpringerInternationalPublishingAG partofSpringerNature Theregisteredcompanyaddressis:Gewerbestrasse11,6330Cham,Switzerland Preface Since the inception offuzzy sets, fuzzy pattern recognition, including its method- ology, algorithms, and applications, has been at the center of the developments of the technology of fuzzy sets. One can refer here to the seminal paper entitled Abstraction and pattern classification authored by Bellman, Kalaba, and Zadeh, whichhasopenedunchartedresearchareasandofferedanewattractiveinsightinto theprinciplesofpatternclassification.Asofnow,patternrecognitionaugmentedby the methodology and algorithms of fuzzy sets has established itself as a mature, well-developedresearchdiscipline with avariety ofadvanced applications.Pattern recognition comes with a great deal of challenges, exhibits a continuous paradigm shift (quite often dictated by new applications), and becomes vividly manifested through a growing diversity of areas of its usage. All of those call for substantial enhancements of the existing fundamentals or the formation of new paradigms. Computational Intelligence (CI) with its impressive armamentarium of methodologiesandtoolsispositionedinauniquewaytoaddressthegrowingneeds of pattern recognition. As a matter of fact, this can be accomplished in several tangible ways realized both at the methodological and algorithmic level. There are atleastfivedominantmanifestationsofCIintherealmofpatternrecognition.They are associated with: (i) coping with a large volume of data and their diversity, (ii) setting a suitable level of abstraction, (iii) dealing with a distributed nature of data along with associated requirements of privacy and security, (iv) building efficient featurespaces, and (v) building interpretable findings ofclassification ata suitable level of abstraction. The key objective of the proposed volume is to provide the community with a comprehensive and up-to-date treatise in the area of pattern recognition and com- putational intelligence. It covers a spectrum of methodological and algorithmic issues, discusses implementations and case studies, identifies the best design practices, and assesses business models and practices of pattern recognition in industry, health care, administration, and business. The collection of contributions forming the edited volume offers the reader a representative view at the progress and accomplishments of the area with a timely, in-depth, and comprehensive v vi Preface material on the conceptually appealing and practically sound methodology and practices of CI-based pattern recognition. ThebookengagesawealthofmethodsofCI,bringsnewconcepts,architectures andpracticeoffuzzysets,neurocomputing,andbiologicallyinspired optimization. The chapters cover a wealth of ideas, algorithms, and applications and are a tes- timonytothesynergisticlinkageswithintheCIareaandCIandpatternrecognition. Given the leading theme of this undertaking, the book is aimed at a broad audienceofresearchersandpractitioners.Thankstothenatureofthematerialbeing coveredandthewaythemainthreadshavebeenorganized,thevolumewillappeal tothewell-establishedcommunitiesincludingthoseactiveinvariousdisciplinesin which pattern recognition plays a central role and serves as an efficient vehicle to produce solutions to numerous classification problems and augments solutions constructed with the aid of the “standard” methodology and algorithms of pattern recognition. With the required prerequisites covered, the book caters to the broad readership. Those involved in operations research, management, various branches of engineering, sciences, data science, medicine, and bioinformatics will benefit from the exposure to the subject matter. We would like to take this opportunity to express our sincere thanks to the contributors to the volume for sharing results of their advanced, far-reaching, and original research, and delivering their views at the rapidly expanding areas of fundamental and applied research. The reviewers deserve our thanks for their constructive and timely input. We greatly appreciate a continuous support and encouragement coming from the Editor-in-Chief, Prof. Janusz Kacprzyk whose leadership and vision has helped us arrive at the successful completion of this project.Theeditorial staffatSpringerhasdoneameticulous jobandworkingwith them was a pleasant experience. Wehopethatthereaderswillfindthisvolumeinterestingandthevarietyofideas put forward in this volume will become instrumental in fostering the progress in research, education, and numerous practical endeavors in the CI-oriented pattern recognition. Edmonton, Canada Witold Pedrycz Taipei, Taiwan Shyi-Ming Chen Contents Fuzzy Choquet Integration of Deep Convolutional Neural Networks for Remote Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Derek T. Anderson, Grant J. Scott, Muhammad Aminul Islam, Bryce Murray and Richard Marcum Deep Neural Networks for Structured Data. . . . . . . . . . . . . . . . . . . . . . 29 Monica Bianchini, Giovanna Maria Dimitri, Marco Maggini and Franco Scarselli Granular Computing Techniques for Bioinformatics Pattern Recognition Problems in Non-metric Spaces . . . . . . . . . . . . . . . . . . . . 53 Alessio Martino, Alessandro Giuliani and Antonello Rizzi Multi-classifier-Systems: Architectures, Algorithms and Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Peter Bellmann, Patrick Thiam and Friedhelm Schwenker Learning Label Dependency and Label Preference Relations in Graded Multi-label Classification. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Khalil Laghmari, Christophe Marsala and Mohammed Ramdani Improving Sparse Representation-Based Classification Using Local Principal Component Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 Chelsea Weaver and Naoki Saito Robust Constrained Concept Factorization . . . . . . . . . . . . . . . . . . . . . . 207 Wei Yan and Bob Zhang An Automatic Cycling Performance Measurement System Based on ANFIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 Andre Vieira Pigatto and Alexandre Balbinot vii viii Contents Fuzzy Classifiers Learned Through SVMs with Application to Specific Object Detection and Shape Extraction Using an RGB-D Camera. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Chia-Feng Juang and Guo-Cyuan Chen Particle Swarm Optimization Based HMM Parameter Estimation for Spectrum Sensing in Cognitive Radio System . . . . . . . . . . . . . . . . . . . . 275 Yogesh Vineetha, E. S. Gopi and Shaik Mahammad Computational Intelligence for Pattern Recognition in EEG Signals . . . 291 Aunnoy K Mutasim, Rayhan Sardar Tipu, M. Raihanul Bashar, Md. Kafiul Islam and M. Ashraful Amin Neural Network Based Physical Disorder Recognition for Elderly Health Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 Sriparna Saha and Raktim Das Recognizing Subtle Micro-facial Expressions Using Fuzzy Histogram of Optical Flow Orientations and Feature Selection Methods. . . . . . . . . 341 S. L. Happy and Aurobinda Routray Improved Deep Neural Network Object Tracking System for Applications in Home Robotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369 Berat A. Erol, Abhijit Majumdar, Jonathan Lwowski, Patrick Benavidez, Paul Rad and Mo Jamshidi Low Cost Parkinson’s Disease Early Detection and Classification Based on Voice and Electromyography Signal . . . . . . . . . . . . . . . . . . . . 397 FarikaT.Putri,MochammadAriyanto,WahyuCaesarendra,RifkyIsmail, Kharisma Agung Pambudi and Elta Diah Pasmanasari Index .... .... .... .... .... ..... .... .... .... .... .... ..... .... 427 Fuzzy Choquet Integration of Deep Convolutional Neural Networks for Remote Sensing DerekT.Anderson,GrantJ.Scott,MuhammadAminulIslam, BryceMurrayandRichardMarcum Abstract Whatdeeplearninglacksatthemomentistheheterogeneousanddynamic capabilities of the human system. In part, this is because a single architecture is not currently capable of the level of modeling and representation of the complex human system. Therefore, a heterogeneous set of pathways from sensory stimu- lus to cognitive function needs to be developed in a richer computational model. Herein,weexplorethelearningofmultiplepathways–asdifferentdeepneuralnet- work architectures–coupled with appropriate data/information fusion. Specifically, weexploretheadvantageofdata-drivenoptimizationoffusingdifferentdeepnets– GoogleNet, CaffeNet and ResNet–at a per class (neuron) or shared weight (single data fusion across classes) fashion. In addition, we explore indices that tell us the importance of each network, how they interact and what aggregation was learned. ExperimentsareprovidedinthecontextofremotesensingontheUCMercedand WHU-RS19datasets.Inparticular,weshowthatfusionisthetopperformer,each network is needed across the various target classes, and unique aggregations (i.e., notcommonoperators)arelearned. · · Keywords Fuzzyintegral Convolutionalneuralnetwork Remotesensing B D.T.Anderson( )·G.J.Scott·R.Marcum ElectricalEngineeringandComputerScience,UniversityofMissouri,Columbia, MO,USA e-mail:[email protected] G.J.Scott e-mail:[email protected] R.Marcum e-mail:[email protected] M.A.Islam·B.Murray ElectricalandComputerEngineering,MississippiStateUniversity, Starkville,MS,USA e-mail:[email protected] B.Murray e-mail:[email protected] ©SpringerInternationalPublishingAG,partofSpringerNature2018 1 W.PedryczandS.-M.Chen(eds.),ComputationalIntelligence forPatternRecognition,StudiesinComputationalIntelligence777, https://doi.org/10.1007/978-3-319-89629-8_1
The list of books you might like
