Embed Size (px)
Citation preview
Proceedings
2014 International Conference on
Identification, Information and Knowledge
in the Internet of Things
IIKI 2014
17-18 October 2014
Beijing, China
Other Sponsors Beijing Normal University
National Natural Science Foundation of China IEEE Communications Society Emerging Technical Committee in Internet of Things
College of Information Science and Technology, Beijing Normal University Research Group on the Internet of Things, Beijing Normal University
Chinese Institute of Electronics Manifesto Group Brunel University
Los Alamitos, California
Washington • Tokyo
Copyright © 2014 by The Institute of Electrical and Electronics Engineers, Inc.
All rights reserved.
Copyright and Reprint Permissions: Abstracting is permitted with credit to the source. Libraries may photocopy
beyond the limits of US copyright law, for private use of patrons, those articles in this volume that carry a code at
the bottom of the first page, provided that the per-copy fee indicated in the code is paid through the Copyright
Clearance Center, 222 Rosewood Drive, Danvers, MA 01923.
Other copying, reprint, or republication requests should be addressed to: IEEE Copyrights Manager, IEEE Service
Center, 445 Hoes Lane, P.O. Box 133, Piscataway, NJ 08855-1331.
The papers in this book comprise the proceedings of the meeting mentioned on the cover and title page. They
reflect the authors’ opinions and, in the interests of timely dissemination, are published as presented and without
change. Their inclusion in this publication does not necessarily constitute endorsement by the editors, the IEEE
Computer Society, or the Institute of Electrical and Electronics Engineers, Inc.
IEEE Computer Society Order Number E5402
BMS Part Number CFP14B18-ART
ISBN-13: 978-1-4799-8003-1
Additional copies may be ordered from:
IEEE Computer Society IEEE Service Center IEEE Computer Society
Customer Service Center 445 Hoes Lane Asia/Pacific Office
10662 Los Vaqueros Circle P.O. Box 1331 Watanabe Bldg., 1-4-2
P.O. Box 3014 Piscataway, NJ 08855-1331 Minami-Aoyama
Los Alamitos, CA 90720-1314 Tel: + 1 732 981 0060 Minato-ku, Tokyo 107-0062
Tel: + 1 800 272 6657 Fax: + 1 732 981 9667 JAPAN
Fax: + 1 714 821 4641 http://shop.ieee.org/store/ Tel: + 81 3 3408 3118
http://computer.org/cspress
[email protected] Fax: + 81 3 3408 3553
Individual paper REPRINTS may be ordered at: <[email protected]>
Editorial production by Randall Bilof
Art production by Mark J. Bartosik
IEEE Computer Society
Conference Publishing Services (CPS) http://www.computer.org/cps
2014 International Conferenceon Identification, Informationand Knowledge in the Internet
of Things
IIKI 2014Table of Contents
Message from the Conference Chairs.......................................................................................................xi
Organizing and Program Committees......................................................................................................xii
Reviewers...................................................................................................................................................xiv
KeynotesIoT and the Need for High Performance Computing ....................................................................................1
Didier El Baz
Wearable Internet: Powering Personal Devices with the Internet of ThingsCapabilities ...................................................................................................................................................7
Antonio J. Jara
Distortion-driven Turbulence Image Blur Effect Removal using VariationalModel and Kernel Regression ......................................................................................................................8
Wensheng Zhang
Session 1: Knowledge Engineering, Big Data, and CloudComputingAn Anti-theft Electric Bicycle Tracking System Supporting Large-Scale Users ............................................9
Jun Zeng, Minbo Li, and Jia Liang
Chinese Social Media Analysis for Disease Surveillance ...........................................................................17Nanhai Yang, Xiaohui Cui, Cheng Hu, Weiping Zhu, and Chengrui Yang
A Novel Dynamic Weight Neural Network Ensemble Model ......................................................................22Kewen Li, Wenying Liu, Kang Zhao, Weishan Zhang, and Lu Liu
Towards a High Speed Video Cloud Based on Batch Processing Integratedwith Fast Processing ...................................................................................................................................28
Weishan Zhang, Liang Xu, Pengcheng Duan, Wenjuan Gong, Xin Liu,and Qinghua Lu
vv
An Improved SMOTE Imbalanced Data Classification Method Basedon Support Degree .....................................................................................................................................34
Kewen Li, Wenrong Zhang, Qinghua Lu, and Xianghua Fang
Simulation and Evaluation of Decentralized SPARQL Query Processing ..................................................39Jing Zhou, Qi Huang, and Wei Yan
Chinese Temporal Relation Resolution Based on Chinese-English ParallelCorpus ........................................................................................................................................................45
Lubiao Li, Junsheng Zhang, Yanqing He, Yinsheng Zhang, and Huilin Wang
Linear Programming v-Nonparallel Support Vector Machine .....................................................................51Guangyu Zhu and Peng Zhang
Big Data Processing: Data Flow vs Control Flow (New BenchmarkingMethodology) ..............................................................................................................................................56
Anton Kos, Sašo Tomažc; Jakob Salom, Nemanja Trifunovic, Mateo Valero,and Veljko Milutinovic
Term Extraction Using Co-occurrence in Abstract and First Claim for PatentAnalysis .......................................................................................................................................................60
Peng Qu, Junsheng Zhang, Yanqing He, Wen Zeng, and Hongjiao Xu
A Short Survey on Decision Making for Task Migrations in Mobile CloudEnvironments ..............................................................................................................................................64
Weishan Zhang, Shouchao Tan, and Klaus Marius Hansen
A Cloud Based Object Recognition Platform for IOS ..................................................................................68Lianzhang Zhu, Xuexing Zheng, Pengfei Li, and Yong Wang
Online Multiperson Tracking and Counting with Cloud Computing ............................................................72Weishan Zhang, Wenshan Wang, Pengcheng Duan, Xin Liu, and Qinghua Lu
Enhancing Context-Aware Recommendation via a Unified Graph Model ..................................................76Hao Wu, Xiaoxin Liu, Yijian Pei, and Bo Li
Online Optimization of Collaborative Web Service QoS Prediction Basedon Approximate Dynamic Programming .....................................................................................................80
Xiong Luo, Hao Luo, and Xiaohui Chang
A Node Localization Approach Using Particle Swarm Optimization in WirelessSensor Networks ........................................................................................................................................84
Xihai Zhang, Tianjian Wang, and Junlong Fang
Enhanced Web Warehouse Model: A Secure Approach ............................................................................88Rashid Mehmood, Maqbool Uddin Shaikh, Liran Ma, and Rongfang Bie
Internet of Things Services for Small Towns ..............................................................................................92Yunchuan Sun, Ye Xia, Houbing Song, and Rongfang Bie
Analysis of Erdös Collaboration Graph and the Paper Citation Network ....................................................96Chengrui Yang, Xiaohui Cui, Xiaoyong Sun, Yuanda Diao, Shuai Wang,and Cheng Hu
vivi
A Relational Model Based Semantic Network Knowledge RepresentationTechnology and Its Application .................................................................................................................100
Yuexin Li and Rong Xiao
Research on Database Massive Data Processing and Mining Methodbased on Hadoop Cloud Platform .............................................................................................................107
Dan Wu, Zhuorong LI, Rongfang Bie, and Mingquan Zhou
Session 2: Pervasive Service Systems and WearableComputingApplication of Inertial Navigation System in Portable Human Body Joint PowerTest System ..............................................................................................................................................111
Lin Li, Zhongqiu Ji, Ye Xia, and Rui Gong
Sensors Classification for Stress Analysis: Toward Automatic StressRecognition ...............................................................................................................................................117
Mikhail Sysoev, Andrej Kos, Urban Sedlar, and Matevž Pogacnik
Autonomous Wearable Personal Training System with Real-Time Biofeedbackand Gesture User Interface ......................................................................................................................122
Anton Umek, Sašo Tomažic, and Anton Kos
Robustness of Input Features from Noisy Silhouettes in Human PoseEstimation .................................................................................................................................................126
Wenjuan Gong, Preben Fihl, Jordi Gonzalez, Thomas B. Moueslund,Weishan Zhang, Zhen Li, and Yan Ren
The Research on Sports Events Organization and Management InformationSystem Based on Process Aware ............................................................................................................132
Yunchao Ma and Zhongqiu Ji
Session 3: Wireless and Mobile SecurityAn Encryption Depth Optimization Scheme for Fully Homomorphic Encryption ......................................137
Liquan Chen, Hongmei Ben, and Jie Huang
Mining Call Spammers from Logs .............................................................................................................142Zhipeng Liu and Weihua Duan
An Improved Multi-sensor Image Fusion Algorithm ..................................................................................146Zhuozheng Wang and John R. Deller
An Online Anomaly Learning and Forecasting Model for Large-Scale Serviceof Internet of Things ..................................................................................................................................152
Junping Wang and Shihui Duan
A Fast AES Encryption Method Based on Single LUT for Industrial WirelessNetwork .....................................................................................................................................................158
Xinqiang Luo, Yue Qi, Yadong Wan, Qin Wang, and Hong Zhang
viivii
Privacy Information Security Classification Study in Internet of Things ....................................................162Xiaofeng Lu, Qi Li, Zhaowei Qu, and Pan Hui
Time Synchronization Attacks in IEEE802.15.4e Networks .....................................................................166Wei Yang, Qin Wang, Yue Qi, and Shaobo Sun
A Fuzzy Operator-Attribute-Based Signcryption Scheme on Vehicular Clouds .......................................170Zhang Wenbo, Yang Pengfei, Bao Zhenshan, Duan Lijuan, and Li Jian
Joint Social and Physical Clustering Scheme for Device-to-DeviceCommunications .......................................................................................................................................175
Chunyan Cao, Li Wang, and Mei Song
Session 4: Frontiers in Cyber-Physical SystemsA Fuzzy Filter for Color Images Corrupted by Mixed Noise ......................................................................177
Xuan Guo and Baoping Guo
Towards a Genetic Algorithm Based Approach for Task Migrations ........................................................182Weishan Zhang, Shouchao Tan, Qinghua Lu, and Xin Liu
Virtual Power Meter Supported Power Consumption Prediction of WebServices ....................................................................................................................................................188
Jiaming Jiang, Jinyang Liu, Lingfeng Wei, Lei Lei, Jing Du, and Jin Liu
The Study of MAC Protocol for Industrial Wireless Sensor Network Basedon Ultra-wide Band ...................................................................................................................................194
Wen Zeng, Junsheng Zhang, and Peng Qu
Diffusion Dynamics in Structured Online Social Networks with Push-BasedForwarding Mechanism ............................................................................................................................198
Pei Li, Fengcai Qiao, Yini Zhang, and Hui Wang
Impact of Structure Balance on Opinion Spreading in Signed Social Networks .......................................202Pei Li, Su He, Yini Zhang, and Hui Wang
Planar Waypoint Generation and Path Finding in Dynamic Environment ................................................206Daoyuan Jia, Cheng Hu, Kechen Qin, and Xiaohui Cui
A Computational Simulation Model for Understanding the Correlationof Climate Change and Population Migration ...........................................................................................212
Cheng Hu, Liang Zhou, Xiaohui Cui, and Yang Zhang
Cache-Based Periodic Query Optimization for Wireless Sensor Networks ..............................................216Deng Zhao, Zhangbing Zhou, Ke Ning, and Xiaolei Wang
An Orthogonal Cartesian Genetic Programming Algorithm for EvolvableHardware ..................................................................................................................................................220
Fuchuan Ni, Yuanxiang Li, Xiaoyan Yang, Fuchuan Ni, and Jinhai Xiang
Detection of Acute Hypotensive Episodes via Empirical Mode Decompositionand Genetic Programming ........................................................................................................................225
Dazhi Jiang, Liyu Li, Zhun Fan, and Jin Liu
viiiviii
A Random Factor Extension on the PSO Algorithm .................................................................................229Huibin Zhang, Jie Lin, Yungang Wei, Lanjun Duan, and Xiaoming Zhu
Subway Fire Evacuation Simulation Model ..............................................................................................233Kechen Qin, Cheng Hu, Daoyuan Jia, Xiaohui Cui, and Yang Zhang
Session 5: Regulations, Standards, User Experience,and Industries in the Internet of ThingsEnterprise-Oriented IoT Name Service for Agriculture Product Supply ChainManagement .............................................................................................................................................237
Yi Liu, He Wang, Junyu Wang, Kan Qian, Ning Kong, Kaijiang Wang, Yiwei Shi,and Lirong Zheng
Multiplex TDMA Link Assignment with Varying Number of Sensors in IndustrialWireless Sensor Networks ........................................................................................................................242
Yanhong Yang and Shaozhong Cao
Patented Network Analysis on Cloud Computing Technology in Internetof Things ...................................................................................................................................................248
Yu Yuan, Li Ma, and Junsheng Zhang
A MapReduce Enabled Simulated Annealing Genetic Algorithm .............................................................252Luokai Hu, Jin Liu, Chao Liang, and Fuchuan Ni
Study of Intelligent Instrument Data Acquisition and Transmission on WirelessNetwork .....................................................................................................................................................256
Huating Fu and Guofeng Qin
Big and Small Data: The Fog ....................................................................................................................260Olga Ferrer-Roca, Ruben Tous, and Rodolfo Milito
Ladle Monitor System Based on Vehicle Distance Measurement and AuxiliaryJudgment Rules ........................................................................................................................................262
Cai Jun and Wang Hong-Bing
Session 6: Mobile Opportunistic NetworksImplementing PMIPv6 Protocol Based on Extended Service Set for IEEE802.11 Infrastructure WLAN .....................................................................................................................266
Deqing Zhu, Lin Zu, Yi-Hua Zhu, and Xianzhong Tian
Spacial Mobility Prediction Based Routing Schemein Delay/Disruption-Tolerant Networks .....................................................................................................274
Lichen Zhang, Zhipeng Cai, Junling Lu, and Xiaoming Wang
Secure Friend Discovery Based on Encounter History in Mobile SocialNetworks ...................................................................................................................................................280
Hongjuan Li, Yingwen Chen, Xiuzhen Cheng, Keqiu Li, and Dechang Chen
ixix
The Dissemination Distance of Mobile Opportunistic Networks ...............................................................286Xia Wang, Shengling Wang, Wenshuang Liang, Rongfang Bie, and Feng Zhao
S-Disjunct Code Based MAC Protocol for Reliable Broadcast in Vehicular AdHoc Networks ...........................................................................................................................................291
Chao Wang, Xiumei Fan, Jiguo Yu, Kai Xing, Yingwen Chen, and Jiawei Liang
Nodes Density Adaptive Opportunistic Forwarding Protocol for IntermittentlyConnected Networks ................................................................................................................................297
Xiaofeng Lu, Pietro Lio, Pan Hui, and Zhaowei Qu
Reliability Evaluation of Coal Mine Internet of Things ..............................................................................301Pan Kunkun and Li Xiangong
Author Index ............................................................................................................................................303
xx
IEEE Computer Society Technical & Conference
Activities Board
T&C Board Vice President
Cecilia Metra Università di Bologna, Italy
IEEE Computer Society Staff
Evan Butterfield, Director of Products and Services Lynne Harris, CMP, Senior Manager, Conference Support Services Patrick Kellenberger, Supervisor, Conference Publishing Services
IEEE Computer Society Publications
The world-renowned IEEE Computer Society publishes, promotes, and distributes a wide variety of authoritative computer science and engineering texts. These books are available from most retail outlets. Visit the CS Store at http://www.computer.org/portal/site/store/index.jsp for a list of products.
IEEE Computer Society Conference Publishing Services (CPS) The IEEE Computer Society produces conference publications for more than 300 acclaimed international conferences each year in a variety of formats, including books, CD-ROMs, USB Drives, and on-line publications. For information about the IEEE Computer Society’s Conference Publishing Services (CPS), please e-mail: [email protected] or telephone +1-714-821-8380. Fax +1-714-761-1784. Additional information about Conference Publishing Services (CPS) can be accessed from our web site at: http://www.computer.org/cps
Revised: 18 January 2012
CPS Online is our innovative online collaborative conference publishing system designed to speed the delivery of price quotations and provide conferences with real-time access to all of a project's publication materials during production, including the final papers. The CPS Online workspace gives a conference the opportunity to upload files through any Web browser, check status and scheduling on their project, make changes to the Table of Contents and Front Matter, approve editorial changes and proofs, and communicate with their CPS editor through discussion forums, chat tools, commenting tools and e-mail.
The following is the URL link to the CPS Online Publishing Inquiry Form: http://www.computer.org/portal/web/cscps/quote
306
Big and Small data The FOG.
Olga Ferrer-RocaUnesco Chair of Telemedicine. Faculty of Medicine.
University of La Laguna. Tenerife. [email protected]
Ruben Tous
Barcelona Supercomputing Center (BSC) and the Universitat Politècnica de Catalunya(UPC) - BarcelonaTech
Barcelona, Spain. [email protected]
Rodolfo Milito Senior Technical Leader CTAO. Fog computing platform.
Cisco Systems. ConSentry networks. [email protected]
Abstract— Health 4.0 applications in the IoE (Internet of Everything)
framework generate and use both “Small Data”, and “Big Data”. While “Big Data” is processed in the Cloud, we advocate for “Small Data” to be processed in the Fog, which is an extension of the Cloud to the edge of the network (close to the IoT devices that stream private health-related information). Processing and storing Small Data close to the sources has enables tighter control of the data ownership, response time, and semi-autonomy require by critical applications.
Keywords—Small data; Health 4.0; mHealth; The Fog; The Cloud.
I. INTRODUCTION
The Global Health 2035 vision calls for Digital and Ubiquitous Health, and Equitable access. Realization of the vision requires the incorporation of advanced Big Data and Analytics techniques, including Data Mining and Machine Learning, making information the center piece of the advances. Natural questions emerge regarding the preservation of privacy, and keeping the control of the information in the hands of the patient. Towards these goals this paper proposes that the data generated by wearables and personal devices be processed and stored in the Fog.
II. CONCEPTS
A. Small dataSmall Data refers to patient data. The patient, who is the
owner of his/her data, must be in control of what is shared,with whom, for what purpose, and during which period. In general this data come from devices (medical or not) connected to Internet as part of the IoT (Internet of the Things) providing information that supports data of interest conducive to a healthy life.
B. Health 4.0As defined in 2012 [1] Health 4.0 is the integrated health
framework that incorporates four main innovations:1º.- Applications that meet three availability criteria: a) Anytime connections: On the move, indoors and outdoors,
day-&-night. b)Anyplace connection: On the move, outdoors, indoors, at any PC. c) Anything connection: At any PC, H2H (human to human), H2T (human to thing), T2T (thing to thing). 2º.- Applications that include image enhancement & RFID readings to be use for: a) People byfaces recognition and access to relevant information (home,work, medical, HER, PHR, medical schedule…). b) Objectby use and by owner recognition. c) Food by principle content & by diet requirements. d) Medication by principle & by indication-contraindication. 3º.-Application that includes quality controlled Web 3.0 items such as: a) HCQ Health Care Quality: ISO 13485-ISO 2700 or security. b) 3S: Social-Semantic-Services. c) Cloud accessing (SAAS, pCloud or personal Cloud were the iPhone can be included).4º.-Applications taking Web 4.0 items such as: a) KBL o Knowledge base learning, including literature base learning (LBL), Evidence Based learning (EBL), trial base learning (TBL), Image based learning (IBL) etc…b) QBE o Query by example, including query by image (QBI) etc…c) CoLD or Cloud of link data with Artificial intelligence.”
C. The FogFog [3] extends Cloud resources (processing, compute,
and networking) to the edge of the network. Through virtualization, the Fog enables the user to control his/her own data. The user determines the engagement policies of his/her PHA (personal health assistant), including what and with whom to share data, whom to associate (possibly including electric PHAs, e-PHAs) with the purpose tomaintain a healthy life.
D. PHAsThe living Personal Health Assistants are highly trained
nurses (e.g. midwives) capable to be informed and empathic, advice for every-day life, and to act as a coach. Their role goes well beyond the standard tasks of prescribing medicines and issuing orders. Nurses that handle the administrative complexity of healthcare delivery can also advise users on the use public and private care, on how to prevent complications, choose the best hospital, call an ambulance and prepare the entrance in emergency rooms, etc. PHAs follow well-designed protocols based on decision trees (“given this and that, take this course”). The protocols are based on Bayesian inference or Markovian models.
2014 International Conference on Identification, Information and Knowledge in the Internet of Things
978-1-4799-8003-1/14 $31.00 © 2014 IEEE
DOI 10.1109/IIKI.2014.60
260
E. Intelli-agents and e-PHAsIntelligent agents, judiciously trained on the patient
Small Data, and rigorously tested and validated on the available Big Data, could eventually graduate to e-PHAs able to give recommendations in specific individual situations. The criterion for graduation should be to meet or exceed the performance of the live PHAs following the existing protocols. Rather than e-PHAs issuing rote recommendations we envision them delivering Knowledge on Demand (KoD) [6], based on the data stored in the Fog, and in respond to the stimuli of the diverse sensors that measure the individual and the environment.
III. SAHA (SMART AGENTS HEALTH ARENA) The components of the system include avatars as
embodiment of people, the Smart Agents Health Arena (SAHA) [4] as the environment in which the avatars and intelligent agents interact with the traditional professionals, medical devices and IoT in general in the health arena,including the Fog [7].
Some of this exists today although in a very primitive Health 2.0 or even analog form (see Fig 1)
Fig. 1. Existing avatars-http://sense.ly/ in http://sco.lt/4pv0KH.
A. Digital HealthDigital data is essential in Digital Health, this is the
reason why we have to define also the type of data, for what is taken and to whom it belongs. See table I [8].
TABLE I. THE FOG AND THE SMALL DATA
BIG DATA SMALL DATABelongs to Government/State Individuals/Patient
Anonymized YES NOEncrypted NO YES
Processed in The Cloud The FogResponse Months / Years Minutes / On time
Obtained from Institutions/EHRs Sensors / At home / PHRa
Processed by DB tools-DBaaS Parallel C/e-AgentsStored By Govern/Distributed By Individuals/ PHAsUseful for Decision makers for Individuals
a. Personal Health Record
B. Global HealthFollowing the Lancet Global Health 2035 [1], our next
generation will require not only to study but also to implement a secure digital health environment and a new way to provide medicine.
This include the possibility of de-localized treatment and follow up by Centers of Excellence spread all over the world, possibility that should be assume not only by private insurance but also by public health.
IV. CONCLUSIONSIn this new Healthcare delivery context empower
citizens in their own health control and delivery in several front-ends: 1) Disease and complications prevention. 2) Healthy and happiest life. 3) Humanize health [10] providing PHAs. 4) Integrate primary and secondary care. 5) Improve healthcare resources. 6) Lower health cost and health demands from tertiary care. 7) Know the center of excellence for specific diseases.
Acknowledgment We want to give a warm acknowledgment to all partner
of the H2O project (Humanization of Healthcare) [10] that working together will be able to bring into the reality Global Health in a more humanitarian model.
References
[1] Global health 2035: a world converging within a generation. The Lancet, Volume 382, Issue 9908, Pages 1898 - 1955, 7 December 2013
[2] O.Ferrer-Roca. "Health 4.0 in the i2i era" Intern J Reliable and Quality E-Healthcare, 1(1), 43-57, January-March 2012; http://www.teide.net/catai/Health%204.0%20HD.pdf .
[3] F.Bonomi, R. Milito, J.Xhu, S.Addepalli, “Fog Computing and its Role in the Internet of Things,”, SIGCOMM 2012, http://conferences.sigcomm.org/sigcomm/2012/paper/mcc/p13.pdf
[4] R. Milito. F. Bonomi and P. Monclus. Architecture for Intelligent agents/avatars in Healthcare. CTECH Forum 2010. (not published)
[5] D. Lake, Milito R., Morrow M. and Vangheese R: Internet of Things: Architectural framework for eHealth security. Journal of ICT, vol.3&4,pp: 301-330, 2014, doi: 10.13052/jicts2245-800X.133
[6] O.Ferrer-Roca, A.Figueredo, K.Franco, A.Cardenas. Telemedicine intelligent Learning. Ontology for agent technology. Trans. on Adv. Res. Jul 2005, Vol 1, N. 2, 46-54
[7] O.Ferrer-Roca. (2014) Fog computing in Health 4.0. http://catai.net/blog/2014/05/fog-computing-in-health-4-0/
[8] O. Ferrer-Roca. (2014) Big Data versus Small Data. http://catai.net/blog/2014/06/big-data-versus-small-data/
[9] O. Ferrer-Roca. (2014) Artificial Intelligence in Health 4.0 http://sco.lt/6ntTyz
[10] O. Ferrer-Roca (2014) Humanization of the Healthcare. Moving into the right direction. http://sco.lt/5BfRdh
[11] R. Tous, J. Delgado, T. Zinkl, P. Toran, G. Alcalde, M. Goetz and O. Ferrer-Roca. The Anatomy of an Optical Biopsy Semantic Retrieval System. IEEE Multimedia. April-June 2012 (vol. 19 no. 2). pp. 16-27.
261
