Vikrant Bhateja Amit Joshi Editors Proceedings of the ...P.B. Mane, Savitribai Phule Pune University, Pune, India Rashmi Agarwal, Manav Rachna International University, Faridabad,

Advances in Intelligent Systems and Computing 469

Suresh Chandra SatapathyVikrant BhatejaAmit Joshi Editors

Proceedings of the International Conference on Data Engineering and Communication TechnologyICDECT 2016, Volume 2

Advances in Intelligent Systems and Computing

Volume 469

Series editor

Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Polande-mail: [email protected]

About this Series

The series “Advances in Intelligent Systems and Computing” contains publications on theory,applications, and design methods of Intelligent Systems and Intelligent Computing. Virtuallyall disciplines such as engineering, natural sciences, computer and information science, ICT,economics, business, e-commerce, environment, healthcare, life science are covered. The listof topics spans all the areas of modern intelligent systems and computing.

The publications within “Advances in Intelligent Systems and Computing” are primarilytextbooks and proceedings of important conferences, symposia and congresses. They coversignificant recent developments in the field, both of a foundational and applicable character.An important characteristic feature of the series is the short publication time and world-widedistribution. This permits a rapid and broad dissemination of research results.

Advisory Board

Chairman

Nikhil R. Pal, Indian Statistical Institute, Kolkata, Indiae-mail: [email protected]

Members

Rafael Bello, Universidad Central “Marta Abreu” de Las Villas, Santa Clara, Cubae-mail: [email protected]

Emilio S. Corchado, University of Salamanca, Salamanca, Spaine-mail: [email protected]

Hani Hagras, University of Essex, Colchester, UKe-mail: [email protected]

László T. Kóczy, Széchenyi István University, Győr, Hungarye-mail: [email protected]

Vladik Kreinovich, University of Texas at El Paso, El Paso, USAe-mail: [email protected]

Chin-Teng Lin, National Chiao Tung University, Hsinchu, Taiwane-mail: [email protected]

Jie Lu, University of Technology, Sydney, Australiae-mail: [email protected]

Patricia Melin, Tijuana Institute of Technology, Tijuana, Mexicoe-mail: [email protected]

Nadia Nedjah, State University of Rio de Janeiro, Rio de Janeiro, Brazile-mail: [email protected]

Ngoc Thanh Nguyen, Wroclaw University of Technology, Wroclaw, Polande-mail: [email protected]

Jun Wang, The Chinese University of Hong Kong, Shatin, Hong Konge-mail: [email protected]

More information about this series at http://www.springer.com/series/11156

Suresh Chandra Satapathy ⋅ Vikrant BhatejaAmit JoshiEditors

Proceedings of theInternational Conferenceon Data Engineering andCommunication TechnologyICDECT 2016, Volume 2

123

EditorsSuresh Chandra SatapathyDepartment of CSEAnil Neerukonda Institute of Technologyand Sciences

Visakhapatnam, Andhra PradeshIndia

Vikrant BhatejaShri Ramswaroop Memorial Group ofProfessional Colleges (SRMGPC)

Lucknow, Uttar PradeshIndia

Amit JoshiSabar Institute of TechnologyTajpur, Sabarkantha, GujaratIndia

ISSN 2194-5357 ISSN 2194-5365 (electronic)Advances in Intelligent Systems and ComputingISBN 978-981-10-1677-6 ISBN 978-981-10-1678-3 (eBook)DOI 10.1007/978-981-10-1678-3

Library of Congress Control Number: 2016944918

© Springer Science+Business Media Singapore 2017This 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 Science+Business Media Singapore Pte Ltd.

Preface

The First International Conference on Data Engineering and CommunicationTechnology (ICDECT 2016) was successfully organized by Aspire ResearchFoundation, Pune during March 10–11, 2016 at Lavasa City, Pune. The conferencehas technical collaboration with Div-V (Education and Research) of ComputerSociety of India. The objective of this international conference was to provideopportunities for the researchers, academicians, industry persons, and students tointeract and exchange ideas, experience and gain expertise in the current trends andstrategies for information and intelligent techniques. Research submissions in vari-ous advanced technology areas were received and after a rigorous peer-reviewprocess with the help of program committee members and external reviewer, 160papers in separate two volumes (Vol-I: 80, Vol-II: 80) were accepted. All the papersare published in Springer AISC series. The conference featured seven special ses-sions on various cutting-edge technologies which were conducted by eminent pro-fessors. Many distinguished personalities like Dr. Ashok Deshpande, FoundingChair: Berkeley Initiative in Soft Computing (BISC)—UC Berkeley CA; GuestFaculty: University of California Berkeley; Visiting Professor: New South WalesUniversity, Canberra and Indian Institute of Technology Bombay, Mumbai, India;Dr. Parag Kulkarni, Pune; Prof. Amit Joshi, Sabar Institute, Gujarat; Dr. SwagatamDas, ISI, Kolkata, graced the event and delivered talks on cutting-edge technologies.

Our sincere thanks to all Special Session Chairs (Dr. Vinayak K. Bairagi, Prof.Hardeep Singh, Dr. Divakar Yadav, Dr. V. Suma), Track Manager (Prof. StevenLawrence Fernandes) and distinguished reviewers for their timely technical support.Thanks are due to ASP and its dynamic team members for organizing the event in asmooth manner. We are indebted to Christ Institute of Management for hosting theconference in their campus. Our entire organizing committee, staff of CIM, studentvolunteers deserve a big pat for their tireless efforts to make the event a grandsuccess. Special thanks to our Program Chairs for carrying out an immaculate job.We would like to extend our special thanks here to our publication chairs doing agreat job in making the conference widely visible.

v

Lastly, our heartfelt thanks to all authors without whom the conference wouldnever have happened. Their technical contributions to make our proceedings richare praiseworthy. We sincerely expect readers will find the chapters very useful andinteresting.

Visakhapatnam, India Suresh Chandra SatapathyLucknow, India Vikrant BhatejaTajpur, India Amit Joshi

vi Preface

Organizing Committee

Honorary Chair

Prof. Sanjeevi Kumar Padmanaban

Organizing Committee

Mr. Satish JawaleMr. Abhisehek DhawanMr. Ganesh KhedkarMayura Kumbhar

Program Committee

Prof. Hemanth KumbharProf. Suresh Vishnudas Limkar

Publication Chair

Prof. Vikrant Bhateja, SRMGPC, Lucknow

Publication Co-Chair

Mr. Amit Joshi, CSI Udaipur Chapter

vii

Technical Review Committee

Le Hoang Son, Vietnam National University, Hanoi, VietnamNikhil Bhargava, CSI ADM, Ericsson IndiaKamble Vaibhav Venkatrao, P.E.S. Polytechnic, IndiaArvind Pandey, MMMUT, Gorakhpur (U.P.), IndiaDac-Nhuong Le, VNU University, Hanoi, VietnamFernando Bobillo Ortega, University of Zaragoza, SpainChirag Arora, KIET, Ghaziabad (U.P.), IndiaVimal Mishra, MMMUT, Gorakhpur (U.P.), IndiaSteven Lawrence Fernandes, Sahyadri College of Engineering and ManagementP.B. Mane, Savitribai Phule Pune University, Pune, IndiaRashmi Agarwal, Manav Rachna International University, Faridabad, IndiaKamal Kumar, University of Petroleum and Energy Studies, DehradunHai V. Pham, Hanoi University of Science and Technology, VietnamS.G. Charan, Alcatel-Lucent India Limited, BangaloreFrede Blaabjerg, Aalborg University, DenmarkDeepika Garg, Amity University, Haryana, IndiaBharat Gaikawad, Vivekanand College campus, Aurangabad, IndiaParama Bagchi, MCKV Institute of Engineering, Kolkata, IndiaRajiv Srivastava, Scholar tech education, IndiaVinayak K. Bairagi, AISSMS Institute of Information Technology, Pune, IndiaRakesh Kumar Jha, Shri Mata Vaishnodevi University, Katra, IndiaSergio Valcarcel, Technical University of Madrid, SpainPramod Kumar Jha, Centre for Advanced Systems (CAS), DRDO, IndiaChung Le, Duytan University, Da Nang, VietnamV. Suma, Dayananda Sagar College of Engineering, Bangalore, IndiaUsha Batra, ITM University, Gurgaon, IndiaSourav De, University Institute of Technology, Burdwan, IndiaAnkur Singh Bist, KIET, Ghaziabad, IndiaAgnieszka Boltuc, University of Bialystok, Poland.Anita Kumari, Lovely Professional University, Jalandhar, IndiaM.P. Vasudha, Jain University Bangalore, IndiaSaurabh Maheshwari, Government Women Engineering College, Ajmer, IndiaDhruba Ghosh, Amity University, Noida, IndiaSumit Soman, C-DAC, Noida, IndiaRamakrishna Murthy, GMR Institute of Technology, A.P., IndiaRamesh Nayak, Shree Devi Institute of Technology, Mangalore, India

viii Organizing Committee

Contents

Experimental Analysis on Big Data in IOT-Based Architecture . . . . . . . 1Anupam Bera, Anirban Kundu, Nivedita Ray De Sarkarand De Mou

Morphology Based Approach for Number Plate Extraction . . . . . . . . . . 11Chetan Pardeshi and Priti Rege

NeSeDroid—Android Malware Detection Based on NetworkTraffic and Sensitive Resource Accessing . . . . . . . . . . . . . . . . . . . . . . . . . 19Nguyen Tan Cam and Nguyen Cam Hong Phuoc

RiCoBiT—Ring Connected Binary Tree: A Structuredand Scalable Architecture for Network-on-Chip Based Systems:an Exclusive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31V. Sanju and Niranjan Chiplunkar

Application of Compressed Sensing (CS) for ECG SignalCompression: A Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53Yuvraj V. Parkale and Sanjay L. Nalbalwar

Tracking Pointer Based Approach for Iceberg Query Evaluation. . . . . . 67Kale Sarika Prakash and P.M.J. Pratap

Performance Evaluation of Shortest Path Routing Algorithmsin Real Road Networks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77Nishtha Kesswani

An Outlook in Some Aspects of Hybrid Decision Tree ClassificationApproach: A Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85Archana Panhalkar and Dharmpal Doye

Content Search Quaternary Look-Up Table Architecture . . . . . . . . . . . . 97D.P. Borkute, P.K. Dakhole and Nayan Kumar Nawre

ix

Exhaust Gas Emission Analysis of Automotive VehiclesUsing FPGA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109Ratan R. Tatikonda and Vinayak B. Kulkarni

A Graph-Based Active Learning Approach Using Forest Classifierfor Image Retrieval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119Shrikant Dhawale, Bela Joglekar and Parag Kulkarni

Comparative Analysis of Android Malware Detection Techniques . . . . . 131Nishant Painter and Bintu Kadhiwala

Developing Secure Cloud Storage System Using Access ControlModels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141S.A. Ubale, S.S. Apte and J.D. Bokefode

ETLR—Effective DWH Design Paradigm . . . . . . . . . . . . . . . . . . . . . . . . . 149Sharma Sachin and Kumar Kamal

Prediction of Reactor Performance in CATSOL-Based SulphurRecovery Unit by ANN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159Gunjan Chhabra, Aparna Narayanan, Ninni Singhand Kamal Preet Singh

A Multilevel Clustering Using Multi-hop and Multiheadin VANET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171G. Shanmugasundaram, P. Thiyagarajan, S. Tharaniand R. Rajavandhini

Patient-Specific Cardiac Computational Modeling Based on LeftVentricle Segmentation from Magnetic Resonance Images . . . . . . . . . . . 179Anupama Bhan, Disha Bathla and Ayush Goyal

A Cryptographic Key Generation on a 2D GraphicsUsing RGB Pixel Shuffling and Transposition . . . . . . . . . . . . . . . . . . . . . 189Londhe Swapnali, Jagtap Megha, Shinde Ranjeet, P.P. Belsareand Gavali B. Ashwini

Methods for Individual and Group Decision MakingUsing Interval-Valued Fuzzy Preference Relations . . . . . . . . . . . . . . . . . . 197B.K. Tripathy, Viraj Sahai and Neha Kaushik

A New Approach to Determine Tie-Line Frequency Bias (B)in Interconnected Power System with Integral ControlAGC Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207Charudatta Bangal

Significance of Frequency Band Selection of MFCCfor Text-Independent Speaker Identification . . . . . . . . . . . . . . . . . . . . . . . 217S.B. Dhonde and S.M. Jagade

x Contents

Ensuring Performance of Graphics Processing Units:A Programmer’s Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225Mayank Varshney, Shashidhar G. Koolagudi, Sudhakar Velusamyand Pravin B. Ramteke

Analytical Study of Miniaturization of Microstrip Antennafor Bluetooth/WiMax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237Pranjali Jumle and Prasanna Zade

Novel All-Optical Encoding and Decoding Schemefor Code Preservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245Varinder Kumar Verma, Ashu Verma, Abhishek Sharmaand Sanjeev Verma

XPM-Based Bandwidth Efficient WDM-to-OTDM ConversionUsing HNLF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253Abhishek Sharma and Sushil Kakkar

Analysis of a Sporting Event on a Social Network:True Popularity & Popularity Bond . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261Anand Gupta, Nitish Mittal and Neeraj Kohli

Performance Analysis of LPC and MFCC Features in VoiceConversion Using Artificial Neural Networks . . . . . . . . . . . . . . . . . . . . . . 275Shashidhar G. Koolagudi, B. Kavya Vishwanath, M. Akshathaand Yarlagadda V.S. Murthy

Person Detection and Tracking Using Sparse Matrix Measurementfor Visual Surveillance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281Moiz Hussain and Govind Kharat

Improvisation in Frequent Pattern Mining Technique . . . . . . . . . . . . . . . 295Sagar Gajera and Manmay Badheka

Design and Simulation of Hybrid SETMOS OperatorUsing Multiple Value Logic at 120 nm Technology . . . . . . . . . . . . . . . . . 305Raut Vaishali and P.K. Dakhole

Detailed Survey on Attacks in Wireless Sensor Network . . . . . . . . . . . . . 319A.R. Dhakne and P.N. Chatur

Comparative Analysis of Frontal Face Recognition Using RadialCurves and Back Propagation Neural Network . . . . . . . . . . . . . . . . . . . . 333Latasha Keshwani and Dnyandeo Pete

Data Perturbation: An Approach to Protect Confidential Datain Cloud Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345Dipali Darpe and Jyoti Nighot

Contents xi

Biologically Inspired Techniques for Cognitive Decision-Making . . . . . . 353Ashish Chandiok and D.K. Chaturvedi

Analysis of Edge Detection Techniques for Side Scan SonarImage Using Block Processing and Fuzzy Logic Methods . . . . . . . . . . . . 363U. Anitha and S. Malarkkan

Leveraging Virtualization for Optimal Resource Managementin a Cloud Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371Dhrub Kumar and Naveen Kumar Gondhi

Reconfigurable Circular Microstrip Patch Antennawith Polarization Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383Prachi P. Vast and S.D. Apte

Issues with DCR and NLSR in Named-Based Routing Protocol . . . . . . . 391Rajeev Goyal and Samta Jain Goyal

Design and Analysis of Quantum Dot Cellular Automata TechnologyBased Reversible Multifunction Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399Waje Manisha Govindrao and K. Pravin Dakhole

Text-Independent Automatic Accent Identification Systemfor Kannada Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411R. Soorajkumar, G.N. Girish, Pravin B. Ramteke, Shreyas S. Joshiand Shashidhar G. Koolagudi

A Study on the Effect of Adaptive Boosting on Performanceof Classifiers for Human Activity Recognition . . . . . . . . . . . . . . . . . . . . . 419Kishor H. Walse, Rajiv V. Dharaskar and Vilas M. Thakare

Toward Improved Performance of Emotion Detection:Multimodal Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431R.V. Darekar and A.P. Dhande

Priority Dissection Supervision for Intrusion Detection in WirelessSensor Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445Ayushi Gupta, Ayushi Gupta, Deepali Virmani and Payal Pahwa

Multi-objective Evolution-Based Scheduling of ComputationalIntensive Applications in Grid Environment . . . . . . . . . . . . . . . . . . . . . . . 457Mandeep Kaur

Selective Encryption Framework for Secure MultimediaTransmission over Wireless Multimedia Sensor Networks. . . . . . . . . . . . 469Vinod B. Durdi, Prahlad T. Kulkarni and K.L. Sudha

Mining Frequent Quality Factors of Software SystemUsing Apriori Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481Jyoti Agarwal, Sanjay Kumar Dubey and Rajdev Tiwari

xii Contents

Algorithm for the Enumeration and Identification of KinematicChains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491Suwarna Torgal

A New Congestion Avoidance and Mitigation MechanismBased on Traffic Assignment Factor and Transit Routingin MANET. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 501Jay Prakash, Rakesh Kumar and J.P. Saini

MRWDPP: Multipath Routing Wormhole Detectionand Prevention Protocol in Mobile Ad Hoc Networks . . . . . . . . . . . . . . . 513Ravinder Ahuja, Vinit Saini and Alisha Banga

Self-coordinating Bus Route System to Avoid Bus Bunching. . . . . . . . . . 523Vishal B. Pattanashetty, Nalini C. Iyer, Abhinanadan Dinkarand Supreeta Gudi

Review on Data Hiding in Motion Vectors and in Intra-PredictionModes for Video Compression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533K. Sridhar, Syed Abdul Sattar and M. Chandra Mohan

Generation of Product Cipher for Secure Encryptionand Decryption Based on Vedic Encryption Ideologyand Using Variable and Multiple Keys . . . . . . . . . . . . . . . . . . . . . . . . . . . 541Vaishnavi Kamat

A Novel Integer Representation-Based Approach for Classificationof Text Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 557S.N. Bharath Bhushan, Ajit Danti and Steven Lawrence Fernandes

Communication Device for Differently Abled People:A Prototype Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 565Rajat Sharma, Vikrant Bhateja, S.C. Satapathy and Swarnima Gupta

Combination of PCA and Contourlets for MultispectralImage Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577Anuja Srivastava, Vikrant Bhateja and Aisha Moin

A Behavioral Study of Some Widely Employed Partitionaland Model-Based Clustering Algorithms and Their Hybridizations . . . . 587D. Raja Kishor and N.B. Venkateswarlu

An Adaptive MapReduce Scheduler for Scalable HeterogeneousSystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 603Mohammad Ghoneem and Lalit Kulkarni

Enhancement in Connectivity by Distributed Beamformingin WSN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 613Vandana Raj and Kulvinder Singh

Contents xiii

Sparse Representation Based Query Classification Using LDA TopicModeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621Indrani Bhattacharya and Jaya Sil

Multiple Home Automation on Raspberry Pi . . . . . . . . . . . . . . . . . . . . . . 631Krishna Chaitanya, G. Karudaiyar, C. Deepakand Sainath Bhumi Reddy

Sentiment Analysis Based on A.I. Over Big Data . . . . . . . . . . . . . . . . . . . 641Saroj Kumar, Ankit Kumar Singh, Priya Singh, Abdul Mutalib Khan,Vibhor Agrawal and Mohd Saif Wajid

Negotiation and Monitoring of Service Level Agreementsin Cloud Computing Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 651S. Anithakumari and K. Chandrasekaran

Impact of Performance Management Process on Print OrganizationalPerformance—In Indian Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 661P. Iyswarya and S. Rajaram

Mobility Aware Path Discovery for Efficient Routing in WirelessMultimedia Sensor Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 673Rachana Borawake-Satao and Rajesh Prasad

Emerging Internet of Things in Revolutionizing Healthcare . . . . . . . . . . 683Poonam Bhagade, Shailaja Kanawade and Mangesh Nikose

Swarm Intelligent WSN for Smart City . . . . . . . . . . . . . . . . . . . . . . . . . . 691Shobha S. Nikam and Pradeep B. Mane

Representing Natural Language Sentences in RDF Graphsto Derive Knowledge Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 701S. Murugesh and A. Jaya

A Framework to Enhance Security for OTP SMS in E-BankingEnvironment Using Cryptography and Text Steganography . . . . . . . . . . 709Ananthi Sheshasaayee and D. Sumathy

Study on the Use of Geographic Information Systems (GIS)for Effective Transport Planning for Transport for London (TfL) . . . . . 719M.S. Mokashi, Perdita Okeke and Uma Mohan

An Integration of Big Data and Cloud Computing . . . . . . . . . . . . . . . . . 729Chintureena Thingom and Guydeuk Yeon

A New Approach for Rapid Dispatch to Remote Cooperative Groupswith a Novel Key Archetype Using Voice Authentication . . . . . . . . . . . . 739T.A. Amith, J. Prathima Mabel, Rekha Jayaramand S.M. Bindu Bhargavi

xiv Contents

ICT Enabled Proposed Solutions for Soil Fertility Managementin Indian Agriculture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 749B.G. Premasudha and H.U. Leena

Software Maintenance: From the Perspective of Effortand Cost Requirement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 759Sharon Christa, V. Madhusudhan, V. Suma and Jawahar J. Rao

FPGA Implementation of Low Power Pipelined 32-Bit RISCProcessor Using Clock Gating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 769R. Shashidar, R. Santhosh Kumar, A.M. MahalingaSwamyand M. Roopa

Dynamic Software Aging Detection-Based Fault Tolerant SoftwareRejuvenation Model for Virtualized Environment . . . . . . . . . . . . . . . . . . 779I.M. Umesh and G.N. Srinivasan

Analysis of Group Performance by Swarm Agents in SACAArchitecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789K. Ashwini and M.V. Vijayakumar

Background Modeling and Foreground Object Detection for IndoorVideo Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 799N. Satish Kumar and G. Shobha

Agri-Guide: An Integrated Approach for Plant Disease Precaution,Detection, and Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 809Anjali Chandavale, Suraj Patil and Ashok Sapkal

Determination of Area Change in Water Bodies and Vegetationfor Geological Applications by Using Temporal Satellite Imagesof IRS 1C/1D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 819Mansi Ekbote, Ketan Raut and Yogesh Dandawate

Significance of Independent Component Analysis (ICA)for Epileptic Seizure Detection Using EEG Signals . . . . . . . . . . . . . . . . . 829Varsha K. Harpale and Vinayak K. Bairagi

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 839

Contents xv

About the Editors

Dr. Suresh Chandra Satapathy is currently working as Professor and Head,Department of Computer Science and Engineering at Anil Neerukonda Institute ofTechnology and Sciences (ANITS), Andhra Pradesh, India. He obtained his Ph.D. inComputer Science and Engineering from JNTU Hyderabad and M.Tech. in CSEfrom NIT, Rourkela, Odisha, India. He has 26 years of teaching experience. Hisresearch interests include data mining, machine intelligence, and swarm intelligence.He has acted as program chair of many international conferences and has edited sixvolumes of proceedings from Springer LNCS and AISC series. He is currentlyguiding eight scholars for Ph.D. Dr. Satapathy is also a Senior Member of IEEE.

Prof. Vikrant Bhateja is Professor, Department of Electronics and Communica-tion Engineering, Shri Ramswaroop Memorial Group of Professional Colleges(SRMGPC), Lucknow and also the Head (Academics and Quality Control) in thesame college. His areas of research include digital image and video processing,computer vision, medical imaging, machine learning, pattern analysis and recog-nition, neural networks, soft computing, and bio-inspired computing techniques. Hehas more than 90 quality publications in various international journals and con-ference proceedings. Professor Bhateja has been on TPC and chaired various ses-sions from the above domain in international conferences of IEEE and Springer. Hehas been the track chair and served in the core-technical/editorial teams for thefollowing international conferences: FICTA 2014, CSI 2014 and INDIA 2015under Springer-ASIC Series and INDIACom-2015, ICACCI-2015 under IEEE. Heis associate editor in International Journal of Convergence Computing (IJConvC)and also serving in the editorial board of International Journal of Image Mining(IJIM) under Inderscience Publishers. At present he is guest editor for two specialissues received in International Journal of Rough Sets and Data Analysis (IJRSDA)and International Journal of System Dynamics Applications (IJSDA) under IGIGlobal publications.

xvii

Mr. Amit Joshi has an experience of around 6 years in academic and industry inprestigious organizations of Rajasthan and Gujarat. Currently, he is working asAssistant Professor in Department of Information Technology at Sabar Institute inGujarat. He is an active member of ACM, CSI, AMIE, IEEE, IACSIT-Singapore,IDES, ACEEE, NPA, and many other professional societies. Currently, he isHonorary Secretary of CSI Udaipur Chapter and Honorary Secretary for ACMUdaipur Chapter. He has presented and published more than 40 papers in Nationaland International Journals/Conferences of IEEE, Springer, and ACM. He has alsoedited three books on diversified subjects including Advances in Open SourceMobile Technologies, ICT for Integrated Rural Development, and ICT for Com-petitive Strategies. He has also organized more than 25 national and internationalconferences and workshops including International Conference ETNCC 2011 atUdaipur through IEEE, International Conference ICTCS 2014 at Udaipur throughACM, International Conference ICT4SD 2015 by Springer. He has also served onorganizing and program committees of more than 50 conferences/seminars/workshops throughout the world and presented six invited talks in various con-ferences. For his contribution towards the society, The Institution of Engineers(India), ULC, has given him Appreciation Award on the Celebration of Engineers,2014 and by SIG-WNs Computer Society of India on ACCE, 2012.

xviii About the Editors

Experimental Analysis on Big Datain IOT-Based Architecture

Anupam Bera, Anirban Kundu, Nivedita Ray De Sarkar and De Mou

Abstract In this paper, we are going to discuss about big data processing forInternet of Things (IOT) based data. Our system extracts information withinspecified time frame. Data tracker or interface tracks information directly from bigdata sources. Data tracker transfers data clusters to data controller. Data controllerprocesses each data cluster and makes them smaller after removing possibleredundancies. Big data processing is a challenge to maintain data privacy-relatedprotections. Data controller processes big data clusters and sends them throughsecure and/or hidden channels maintaining data privacy.

Keywords Big data (BD) ⋅ Distributed file system (DFS) ⋅ Hadoop ⋅ Dis-tributed network (DN) ⋅ Internet of things (IOT)

1 Introduction

Internet is a great source for information sharing through hypertext documents,applications on World Wide Web (WWW), electronic mail, telephony andpeer-to-peer networks. Internet is the framework of cooking massive information.Nowadays, a large number of smart devices are connected to Internet, and con-

A. Bera (✉) ⋅ A. Kundu ⋅ N.R. De Sarkar ⋅ De MouNetaji Subhash Engineering College, Kolkata 700152, Indiae-mail: [email protected]

A. Kundue-mail: [email protected]

N.R. De Sarkare-mail: [email protected]

De Moue-mail: [email protected]

A. Bera ⋅ A. Kundu ⋅ N.R. De Sarkar ⋅ De MouComputer Innovative Research Society, Howrah 711103, West Bengal, India

© Springer Science+Business Media Singapore 2017S.C. Satapathy et al. (eds.), Proceedings of the International Conferenceon Data Engineering and Communication Technology, Advances in IntelligentSystems and Computing 469, DOI 10.1007/978-981-10-1678-3_1

1

tinuously transmit massive amount of heterogeneous data. It is a challenging job ofprocessing such huge amount of data in real time. Data should be processed indistributed environment in real time. Data could be stored in offline. Size of data istoo large and distributed, that it requires use of BD analytical tools for processing[1]. There is number of different application domains widely used for datastreaming, storing and processing. These application domains are difficult tomaintain due to size of data and nature of network [2, 3]. User could interact withdatabases to modify, insert, delete, manage information using modern databasesoftware or database management system (DBMS) [4]. BD database activities areincluded in capture, data noise removal, search, sharing, storage, transfer, visual-ization, and information privacy [5].

2 Related Works

BD is a collection of large, multidimensional, heterogeneous datasets that becomesdifficult to manage on typical database management systems. BD refers to databundles whose volumes are beyond the scope of traditional database software tools[5, 6]. It is a useful tool to use predictive analytics or certain advanced methodologyto extract value from data [7]. Typical database management systems failed toprovide storage volumes and efficient data processing due to tremendous growthand massive data volume [5]. In 2012, a researcher “Gartner” has described BD ashigh volume, high velocity and/or high variety information assets which requirenew forms of processing for enhanced decision-making, insight discovery, andprocess optimization [8]. In Hadoop [9] mechanism, JobTrackers [10] have pro-cessed a block of data (δx) from distributed storage followed by analysis of datablocks to generate final results for storing into a data file. Hadoop has executedMapReduce algorithms for data processing in parallel fashion on distinct CPUnodes. Hadoop works as follows: (i) Place job to Hadoop for required process;(ii) Hadoop job client submits job to JobTracker; (iii) JobTracker executes task asper MapReduce implementation; (iv) JobTracker stores data into data file.

3 Proposed Work

Our proposed model has two segments as follows: (i) one side actively receives allsignals from sensors, and forward it to BD database servers; (ii) other side servesrequest from enterprise personnel for data analysis.

System receives query from end-user and system coverts them to smaller size.After conversion of queries, system interface controller has plugged the informationwith minimum loaded interface to fetch data from BD server. Fetched records aresubdivided into multiple data cubes, and further received at controller side for dataanalysis. Data analyzer is used to extract all necessary data from data cubes using

2 A. Bera et al.

two levels of data analyses. Finally, data cubes are received at user end throughsecure data path.

Proposed architecture is shown in Fig. 1. This model has four sections as fol-lows: (i) dynamic query generator; (ii) data retrieval through multiple interface;(iii) two levels of data analyzer; (iv) data transmission through secure channels.

Fig. 1 Proposed architecture of our approach

Experimental Analysis on Big Data in IOT-Based Architecture 3

Dynamic query generator has received user’s key string as input. System hassplitted the string into multiple single word key values. These single word keyvalues are being passed to minimum loaded interface. Interfaces are linked withbackend BD database server, which are large-scale distributed datasets, distributedfile system, e.g. NoSQL database. Structured data as well as unstructured data areavailable in BD system. Data retrieval time using load tracker has been reduced.

Controllers have received large datasets from interface. Controller has startedanalyzing for reducing redundancies after received unstructured datasets from BDsystem. Controllers have worked in two phases such as initially it reduces first-levelredundancies, and then controller applies reduction procedure, as per user requesthaving stringent data analysis when most of the key values are matched with recordsets. Finally, controller breaks these datasets into smaller data cubes to transforminto specific structure for database storage.

These smaller data cubes are passed through secure paths as it is difficult toapply encryption due to large volume of datasets. Few hidden channels have beenintroduced for sending data to end-users. These hidden channels and datacubesequence number are chosen randomly.

Formation of dynamic query generator is one of the most cost effective mech-anisms for selecting datasets from BD system. This is introduced to minimize timecomplexity of total procedure. In this phase, we break user query into number ofmeaningful single word key values. Each user query has been broken, and then keyvalues are sent to minimum loaded interface for retrieving required datasets.

Data analyzer is used to reduce redundancy of datasets and further processescomparatively less error-prone data to send it back to specific user. Two phases ofdata analyses have been performed in our system.

In this model, we have introduced secure transmission of final record sets. It isthe most effective way to enforce data security on BD tools. One of the biggestchallenges in BD industry is to enforce security. Information transmission isassured to users through secured hidden channels. In BD environment, data arebeing floated within distributed network in an unsecured manner. We have intro-duced security measures in BD transmission.

Proposed IOT-based BD model has two application segments. First part receivesdata streams from different synchronous and asynchronous sensors. It is responsiblefor transforming data and storing it into respective BD database servers. The otherpart searches information from BD database.

First segment of proposed IOT-based BD model has three algorithms as follows:(i) Synchronous_Data_Feed_Timer(); (ii) Asynchronous_Data_Feed_Timer();(iii) storeRecord().

Algorithm 1 is a synchronous data stream feeder from different synchronoussensor devices. Each synchronous sensor produces continuous stream. Our algo-rithm has captured continuous stream for transforming into appropriate analyticalformats as mentioned in algorithm (i.e. {timestamp, device id, data stream, alarmtype, remarks}).

Algorithm 1: Synchronous_Data_Feed_Timer ()Input: Device ID, Real time data

4 A. Bera et al.

Output: Time Stamp, Device ID, Data, Alarm type

Step 1:Read devID, dataStep 2:l_thr : = call Find_Lower_Thresold (devID)

/*to get lower threshold of the said device*/

Step 3:h_thr : = call Find_Higher_Thresold (devID)

/*to get higher threshold of the said device*/

Step 4:if l_thr > data; then alarmType = low

/*data to be checked for alarm selection*/

Step 5:else if h_thr < data; then alarmType = highStep 6:else alarmType = withinRangeStep 7:Call storeRecord ({timestamp, devID, data, alarm-Type, remarks})Step 8:Stop

Algorithm 2 scans and collects data stream from asynchronous devices. It isexecuted periodically, either when sensor queue is full or after specified timeintervals. Finally, it passes all received information for processing by data server.

Algorithm 2: Asynchronous_Data_Feed_Timer ()Input: Device ID, dataOutput: Time Stamp, Device ID, Data, Alarm type.

Step 1:if (data < threshold or data > threshold);

then alarmType: = OFR/*OFR-Out Of Range*/

Step 2:else alarmType: = WIR/*WIR-With In Range*/Step 3:Call storeRecord ({timestamp, devID, data, alarm-Type, remarks})Step 4:Stop

Algorithm 3 receives information packs from Algorithm 1 or Algorithm 2, thenconverts all data items into single data string. This string is stored in a data file.Database writer periodically updates string information to database server.

Algorithm 3: storeRecord ()Input: Record setOutput: Output file

Step 1:For each data: record setStep 2:Call concate (recordString, dataItem)Step 3:printfile recordString #fileStep 4:Stop

Experimental Analysis on Big Data in IOT-Based Architecture 5

Second segment of our proposed IOT-based BD model has three algorithms asfollows: (i) Fetch_record(); (ii) First_Level_Query_Analysis();(iii) Second_Level_Query_Analysis().

Algorithm 4 acts as dynamic query generator and raw data retriever from BDsystem. System finds raw data (unstructured data) from distributed datasets in thisphase of processing. It receives query string as input and generates output extractinglarge datasets from BD database.

Algorithm 4: Fetch_Record()Input: Query, level of analysisOutput: Interface id, key value

Step 1:Read query_string, query_levelStep 2:key_values[]: = split_Query_String()Step 3:For each key_values : key_valStep 4:interfaceID : = call minimum_Loaded_Interface()Step 5:record_set[]: = call Find_Record (key_val, interfaceID)Step 6:End For loopStep 7:Return {record_set, query_stringm query_level}Step 8:Stop

Algorithm 5 works for first-level data redundancy minimization. It takes largedatasets as input from Fetch_Record algorithm. Based on user’s query level, eitherdata is transferred for further analysis to BD analyzer, or, transferred for secondlevel reduction.

Algorithm 5: First_Level_Query_Analysis ()Input: Record sets, query string, query levelOutput: Less redundant data

Step 1:For each record_set: recordStep 2:if record (common); then delete recordStep 3:End ForStep 4:if query_level : = high thenStep 5:Call Second_Level_Query_Analysis(record_set, query_string)Step 6:Call Big_Data_Ananlyzer(record_set)Step 7:Stop

Algorithm 6 exhibits final-level reduction before calling BD analyzer. It receivesdata from Algorithm 5 for further reduction.

Algorithm 6: Second_Level_Query_Analysis ()Input: record_set, query_stringOutput: reduced datasets

Step 1:len: = no_of_keyvalues (query_string)Step 2:if count (key_val, record_set) < len/2; then deleterecord

6 A. Bera et al.

Step 3:Call Big_Data_Analyzer (record_set)Step 4:Stop

4 Experimental Results

Following diagrams are exhibited to understand behaviour of our system frame-work in real time. It also shows that our system has a cost effective and timeefficient structure.

In Fig. 2, CPU load has been shown with good performance for executingproposed framework. CPU load is less than 50 % in most of the time. It is animportant part to identify how our model works.

Figure 3 represents real-time performance of proposed BD model with an effecton CPU temperature around 45 °C. Most of the time CPU temperature is near to40 °C.

Figure 4 shows a graph of memory load with respect to time. Primary memoryoccupancy is quite high due to large raw record sets. It is around 90 % of totalmemory available. Less temporary space at run time is occupied by our model.

Our system works in two phases, such as data tracker and controller.In Phase 1, there is ‘I’ number of data trackers responsible for data retrieval and

each successive request has ‘K’ number of key-value pairs. So, we have drawn timeequations of data tracker algorithms as follows:

T Kð Þ=OðlogI KÞ, ð1Þ

where I = number of data trackers; K = number of key-value pairs;

Fig. 2 CPU load distribution in real time

Experimental Analysis on Big Data in IOT-Based Architecture 7

In Phase 2, there is ‘C’ number of controllers responsible for reducing datablocks and each controller receives ‘N’ data blocks from data trackers. Followingequation is the representation of time complexity in controller’s part:

T Nð Þ=OðlogC NÞ, ð2Þ

where C = number of controllers; N = number of data blocks;

5 Conclusion

Information from BD sources has been tracked by data tracker, and subsequentlydata controller has processed the confined information avoiding redundancies.Experimental results have shown CPU load, CPU temperature, and primarymemory load in real time to exhibit characteristics of our proposed system. Cost

Fig. 3 CPU temperature performance with respect to time

Fig. 4 Primary memory load versus time

8 A. Bera et al.

analysis of our system framework is depicted with an aim of achieving betterefficiency.

References

1. Charu C. Aggarwal, Naveen Ashish, Amit Sheth, “The Internet of Things: A Survey from theData-Centric Perspective,” Managing and Mining Sensor Data, Springer, 2012, pp. 383–428.

2. D. Nukarapu, B. Tang, L. Wang, S. Lu, “Data replication in data intensive scientificapplications with performance guarantee,” Parallel and Distributed Systems, IEEETransactions, 2011, pp. 1299–1306.

3. Chi-Jen Wu, Chin-Fu Ku, Jan-Ming Ho, “A Novel Approach for Efficient Big DataBroadcasting,” Knowledge and Data Engineering, IEEE Transactions, 2014, IIS TechnicalReport-12–006.

4. P. Beynon-Davies, “Database Systems,” Palgrave Macmillan, 2004, ISBN 1-4039-1601-2.5. Sugam Sharma, Udoyara S Tim, Johnny Wong, Shashi Gadia, Subhash Sharma, “A Brief

Review on Leading Big Data Models,” Data Science Journal, 2014, Vol. 13.6. M. H. Padgavankar, S. R. Gupta, “Big Data Storage and Challenges,” International Journal of

Computer Science and Information Technologies, Vol. 5, No. 2, 2014, pp. 2218–2223.7. Chris Snijders, Uwe Matzat, Ulf-Dietrich Reips, ““Big Data”: Big gaps of knowledge in the

field of Internet,” International Journal of Internet Science, 2012, Vol. 7, No. 1, pp. 1–5.8. Douglas and Laney, “The importance of ‘Big Data’: A definition”, 2008.9. “Extract, Transform, and Load Big Data with Apache Hadoop,” Intel, Big Data Analytics,

White Paper, 2013.10. “Comparing the Hadoop Distributed File System (HDFS) with the Cassandra File System

(CFS),” Datastax Corporation, 2013.

Experimental Analysis on Big Data in IOT-Based Architecture 9

Morphology Based Approach for NumberPlate Extraction

Chetan Pardeshi and Priti Rege

Abstract Number Plate Recognition identifies vehicle number without humanintervention. It is a computer vision application and it has many important appli-cations. The proposed system consists of two parts: number plate area extractionand character identification. In this paper, morphological operation-based approachis presented for number plate area extraction. Effective segmentation of charactersis done after plate area extraction. Histogram-based character segmentation is asimple and efficient technique used for segmentation. Template matching approachis used for character extraction. Number plate with variable character length poseslimitation on number identification in earlier reported literature. This is taken careof using histogram-based character segmentation method.

Keywords Histogram ⋅ Morphological operations ⋅ Number plate extraction ⋅Template matching ⋅ Thresholding

1 Introduction

Every country has specific vehicle identification system. These systems are used inthe traffic control and surveillance systems, security systems, toll collection at tollplaza and parking assistance system, etc. Human eye can easily recognize thesenumber plates, but designing automated system for this task has many challenges.Blur, unequal illumination, background and foreground color and also many naturalphenomena like rain fall, dust in air may create problem in number extraction. Alsonumber plate standards are different for each country, therefore large number ofvariations are obtained in parameters like, location of number plate, area of numberplate and characters, font and size used for numbers and characters (standard font is

Chetan Pardeshi (✉) ⋅ Priti RegeDepartment of Electronics and Telecommunication, College of Engineering Pune, Pune, Indiae-mail: [email protected]

Priti Regee-mail: [email protected]

© Springer Science+Business Media Singapore 2017S.C. Satapathy et al. (eds.), Proceedings of the International Conferenceon Data Engineering and Communication Technology, Advances in IntelligentSystems and Computing 469, DOI 10.1007/978-981-10-1678-3_2

11

Arial Black), background color (white, yellow or black) and foreground color(black or red), etc., which make the task of number plate extraction difficult.

Number of applications of license plate identification can be listed as parkingassistance facility during ticket collection, unmanned toll collection at toll booths,traffic surveillance system, tracking vehicles during signal violation, vehicle’smarketing research.

Aim of this paper is to implement an efficient method for number plateextraction. Algorithm proposed in this paper identifies characters present on singleline number plate with variable character length.

Rege and Chandrakar [1] has separated text image in document images using runlength searing algorithm and boundary perimeter detection. Morphological opera-tion and bounding box analysis are used by Patel et al. [2]. Owamoyo et al. [3] usedSobel filter along with morphological operations. Algorithm presented by Gilly andRaimond [4] stresses on connected component analysis. Bulugu [5] used edgefinding method to locate the plate in the scene. Kate [6] proposed morphologicaloperation based on area for searching number plate. Kolour [7] has reviewed anumber of license plate detection algorithms and compared their performances inhis paper. His experimentation gives a basis for selection of the most appropriatetechnique for their applications. Parasuraman and Kumar [8] extracted the plateregion using edge detection followed by vertical projection. Proposed algorithm hasfour stages: (i) Acquisition of vehicle image and preprocessing includes conversionof image to gray format, resizing of image, etc. (ii) Marking of area covering thenumber plate in the vehicle image, (iii) Segmentation of characters from the numberplate extracted, and (iv) Recognizing and displaying the segmented characters.

2 Proposed Method for Identification of Letters/Numbersfrom License Plates

This section elucidates the number plate extraction method for single line numberplate. Input to the system is a vehicle image (with clear view of number plate in it)which is captured by digital camera and output is the actual characters present inthat vehicle image. Each character present on input number plate image should atleast have minimum resolution of 24 × 42 pixels and distance of number platefrom camera should be such that it guarantees clear view of numbers present on thenumber plate.

Proposed algorithm consists of following steps:

• Image Preprocessing• Number plate area extraction• Segmentation of each character area in image• Image matching for each character• Output extracted characters in text format

12 Chetan Pardeshi and Priti Rege

2.1 Image Preprocessing

Preprocessing is used to enhance the contrast of the image and for resizing of theimage. RGB image is converted to grayscale image which carries intensity infor-mation. RGB values are converted to grayscale values by forming a weighted sumof the R, G, and B components:

Gray= 0.2989 *R+0.587 *G+0.114 *B

2.1.1 Contrast Enhancement

Captured image may have unevenly distributed lighting or darkness. During edgedetection fine edge details in dark region of the image are eliminated. Also featureedges in bright regions need to be preserved. Top-hat transformation is used topreserve these edge details as well as prominent ones. The main property of top-hatoperator can be applied to contrast enhancement. After applying top-hat filtering,image is converted to binary image from gray scale using Otsu’s algorithm.Figure 1a–d shows various stages of image preprocessing.

2.2 Number Plate Area Detection

System’s speed and accuracy is enhanced using precise number plate area extrac-tion. At this stage, number plate area is extracted from entire preprocessed image.This step reduces processing burden on next stage of identification of numbers fromlicense plate area.

Fig. 1 a Original image, b Gray image, c Top-hat filtered image, d Binary image, e Dilatedimage, f Selected object, g Extracted plate

Morphology Based Approach for Number Plate … 13