Lecture Notes in Networks and Systems 120

Jagdish Chand BansalMukesh Kumar GuptaHarish SharmaBasant Agarwal   Editors

Communication and Intelligent SystemsProceedings of ICCIS 2019

Lecture Notes in Networks and Systems

Volume 120

Series Editor

Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,Warsaw, Poland

Advisory Editors

Fernando Gomide, Department of Computer Engineering and Automation—DCA,School of Electrical and Computer Engineering—FEEC, University of Campinas—UNICAMP, São Paulo, Brazil

Okyay Kaynak, Department of Electrical and Electronic Engineering,Bogazici University, Istanbul, Turkey

Derong Liu, Department of Electrical and Computer Engineering, Universityof Illinois at Chicago, Chicago, USA; Institute of Automation, Chinese Academyof Sciences, Beijing, China

Witold Pedrycz, Department of Electrical and Computer Engineering,University of Alberta, Alberta, Canada; Systems Research Institute,Polish Academy of Sciences, Warsaw, Poland

Marios M. Polycarpou, Department of Electrical and Computer Engineering,KIOS Research Center for Intelligent Systems and Networks, University of Cyprus,Nicosia, Cyprus

Imre J. Rudas, Óbuda University, Budapest, Hungary

Jun Wang, Department of Computer Science, City University of Hong Kong,Kowloon, Hong Kong

The series “Lecture Notes in Networks and Systems” publishes the latestdevelopments in Networks and Systems—quickly, informally and with high quality.Original research reported in proceedings and post-proceedings represents the coreof LNNS.

Volumes published in LNNS embrace all aspects and subfields of, as well as newchallenges in, Networks and Systems.

The series contains proceedings and edited volumes in systems and networks,spanning the areas of Cyber-Physical Systems, Autonomous Systems, SensorNetworks, Control Systems, Energy Systems, Automotive Systems, BiologicalSystems, Vehicular Networking and Connected Vehicles, Aerospace Systems,Automation, Manufacturing, Smart Grids, Nonlinear Systems, Power Systems,Robotics, Social Systems, Economic Systems and other. Of particular value to boththe contributors and the readership are the short publication timeframe and theworld-wide distribution and exposure which enable both a wide and rapiddissemination of research output.

The series covers the theory, applications, and perspectives on the state of the artand future developments relevant to systems and networks, decision making, control,complex processes and related areas, as embedded in the fields of interdisciplinaryand applied sciences, engineering, computer science, physics, economics, social, andlife sciences, as well as the paradigms and methodologies behind them.

** Indexing: The books of this series are submitted to ISI Proceedings,SCOPUS, Google Scholar and Springerlink **

More information about this series at http://www.springer.com/series/15179

Jagdish Chand Bansal • Mukesh Kumar Gupta •

Harish Sharma • Basant AgarwalEditors

Communicationand Intelligent SystemsProceedings of ICCIS 2019

123

EditorsJagdish Chand BansalSouth Asian UniversityNew Delhi, Delhi, India

Mukesh Kumar GuptaSwami Keshvanand Institute of TechnologyManagement & GramothanJaipur, Rajasthan, India

Harish SharmaDepartment of Computer Scienceand EngineeringRajasthan Technical UniversityKota, Rajasthan, India

Basant AgarwalIndian Institute of InformationTechnology, MNIT CampusJaipur, Rajasthan, India

ISSN 2367-3370 ISSN 2367-3389 (electronic)Lecture Notes in Networks and SystemsISBN 978-981-15-3324-2 ISBN 978-981-15-3325-9 (eBook)https://doi.org/10.1007/978-981-15-3325-9

© Springer Nature Singapore Pte Ltd. 2020This 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, expressed or implied, with respect to the material containedherein or for any errors or omissions that may have been made. The publisher remains neutral with regardto jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,Singapore

Organization

General Chairs

Dr. Jagdish Chand Bansal, South Asian University, New Delhi, IndiaDr. Seyedali Mirjalili, Griffith University, BrisbaneProf. (Dr.) Dhirendra Mathur, Professor, RTU, Kota

Technical Program Chairs

Dr. Mahesh Chand Govil, Director, NIT SikkimDr. Nilanjan Dey, Techno India College of Technology, Kolkata

Organizing Chairs

Dr. Mukesh Kumar Gupta, SKIT, JaipurDr. Harish Sharma, RTU, Kota

Organizing Secretaries

Ms. Anjana Sangwan, SKIT, JaipurMr. Kailash Soni, SKIT, Jaipur

v

Coordinators

Dr. S. R. DogiwalDr. Pankaj DhadhichDr. Niketa Sharma

International Advisory Committee

Dr. Alireza Maheri, University of Aberdeen, UKProf. Krishna Busawon, University of Northumbria, NewcastleProf. Ajit Kumar Verma, Western Norway University, NorwayMohammad Shorif Uddin, Jahangirnagar UniversityJoonghoon Kim, University of Korea, South KoreaJunzo Watada, Waseda University, Graduate School of IPSMontaz Ali, University of the WitwatersrandSugam Sharma, Iowa State University, USAProf. Suresh Sundaram, Nanyang Technological University, SingaporeNeel Mani, Dublin City UniversityOleg Monakhov, ICMMG SB RASPhilip Moore, Lanzhou UniversityP. N. Suganthan, Nanyang Technological UniversityAndres Muñoz, Universidad Católica San Antonio de MurciaAndries Engelbrecht, University of PretoriaCarlosa A. Coello Coello, CINVESTAV-IPNH. J. C. Barbosa, Laboratório Nacional de Computação CientíficaDharm Singh, Namibia University of Science and Technology, NamibiaBasant Tiwari, Hawassa University, Awassa, EthiopiaVaibhav Katewa, University of California—Riverside California, USARajalakshmi Selvaraj, College of Science, BIUST, Palapye, BotswanaMohd Helmy Abd Wahab, Universiti Tun Hussein Onn, MalaysiaShireen Panchoo, University of Technology, MauritiusAnicia Peters, NUST, NamibiaRajiv Dhiman, Botho University, BotswanaJanos Arpad Kosa, Pallas Athena University, HungaryXiao-Zhi Gao, Aalto University, FinlandSotirios Ziavras, New Jersey Institute of TechnologyDharm Singh Jat, Namibia University of Science and Technology, Namibia, AfricaPallavi Kaliyar, Department of Mathematics, University of Padua, ItalyMr. Nitin Purohit, Wollo University, Dessie, EthiopiaWei-Chiang Hong, School of Education Intelligent Technology, Jiangsu NormalUniversity, China

vi Organization

Vanita Verma, Panjab University, ChandigarhZ. W. Geem, South KoreaJagdish Chand Bansal, South Asian University, New DelhiNeha Yadav, Korea UniversityAtulya Nagar, Liverpool Hope UniversityKrishna Kambhampaty, North Dakota State University, USADinesh K. Sharma, Professor, Department of Management, Fayetteville StateUniversity, FayettevilleVinod P. Nair, Department of Mathematics, University of Padua, ItalyChhagan Lal, Department of Mathematics, University of Padua, ItalyPiyush Maheshwari, Amity University, Dubai, UAE

Technical Program Committee

Dr. Mahesh Chandra Govil, Director, NIT Sikkim, IndiaAmitava Chatterjee, Jadavpur UniversityDr. Alireza Maheri, University of Aberdeen, UKProf. Krishna Busawon, University of Northumbria, NewcastleDgnanaraj Thomas, Madras Christian CollegeProf. Suresh Sundaram, Nanyang Technological University, SingaporeDr. G. A. Vijayalakshmi Pai, PSG College of TechnologyMunawara Shaik, Indian Institute of Technology (IIT) DelhiNeel Mani, Dublin City UniversityOleg Monakhov, ICMMG SB RASPhilip Moore, Lanzhou UniversityP. K. Kapur, University of DelhiP. N. Suganthan, Nanyang Technological UniversityPunam Bedi, University of DelhiPramodkumar Singh, ABV-IIITM, GwaliorAndres Muñoz, Universidad Católica San Antonio de MurciaAndries Engelbrecht, University of PretoriaCarlosa A. Coello Coello, CINVESTAV-IPND. Ghose, IISc, Bangalore, IndiaD.Nagesh Kumar, Indian Institute of Science, BangaloreProf. Ajit Kumar Verma, Western Norway University, NorwayMohammad Shorif Uddin, Jahangirnagar University, BangladeshGhanshyamsingh Thakur, MANIT, BhopalH. J. C. Barbosa, Laboratório Nacional de Computação CientíficaJoonghoon Kim, University of Korea, South KoreaJunzo Watada, Waseda University, Graduate School of IPSM. K. Tiwari, IIT Kharagpur

Organization vii

Montaz Ali, University of the WitwatersrandSugam Sharma, Iowa State University, USALalit Awasthi, NIT, Jalandhar, PunjabDharm Singh, Namibia University of Science and Technology, Namibia

viii Organization

Preface

The first International Conference on “Communication and Intelligent Systems”(ICCIS 2019) was held at Swami Keshvanand Institute of Technology, Managementand Gramothan (SKIT), Jaipur, Rajasthan, India, on November 9–10, 2019.The ICCIS aims to showcase advanced technologies, techniques, innovations andequipment in computer engineering. It provides a platform for researchers, scholars,experts, technicians, government officials and industry personnel from all over theworld to discuss and share their valuable ideas and experiences. The main keynoteaddresses were given by Dr. Jagdish Chand Bansal, Associate Professor, SouthAsian University, New Delhi; Prof. Nik Mohd Asri Bin Nik Long, Professor,University Putra, Malaysia and Prof. Mahesh Chandra Govil, Director, NIT, Sikkim.

The ICCIS 2019 received 154 submissions from all over the world, includingcountries like Singapore, France, Tunisia, to name a few. The Technical ProgramCommittee members carefully selected the papers after peer review by at least threereviewers. Out of 154 submissions, 39 papers were selected for presentation in theconference and publication in the conference proceedings.

We wish to thank the Honourable Vice Chancellor of Rajasthan TechnicalUniversity, Kota, for financial sponsoring of the event under TEQIP-III RTU(ATU) project. We are also thankful to the management of Swami KeshvanandInstitute of Technology, Management and Gramothan (SKIT), Jaipur, Rajasthan,India, for providing the best infrastructure and required logistics to organize theconference. We wish to thank all the experts and reviewers for all their efforts. Weare also very thankful to Springer for supporting ICCIS 2019 and Mr. Aninda Bose,Dr. Jagdish Chand Bansal and Dr. Dhirendra Mathur for the approval, continuoussupport and help.

We hope that this proceedings would be very useful for the researchers workingin the relevant areas.

Kota, India Dr. Harish SharmaJaipur, India Dr. Mukesh Kumar Gupta

ix

Contents

Extremely High-Capacity, Secure Image SteganographyBased on Composite Edge Detection, with Message Integrity . . . . . . . . . 1Priyanka Yadav and A. K. Mohapatra

A Dynamic Hand Gesture-Based Password Recognition System . . . . . . 21Manasa Pisipati, Amrit Puhan, Arun Kumar, Vijay Bhaskar Semwaland Himanshu Agrawal

Fuzzy Concepts and Machine Learning Algorithms for Car ParkOccupancy and Route Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35Mani V. Sumini, Jaison Mulerikkal, P. B. Ramkumar and Paulsy Tharakan

A Distributed Algorithm to Create a Minimum Spanning Treein a Multi-channel Cognitive Radio Network . . . . . . . . . . . . . . . . . . . . . 53Deepak Rohilla and Mahendra Kumar Murmu

A Distributed Fault Analysis (DFA) Method for Fault Tolerancein High-Performance Computing Systems . . . . . . . . . . . . . . . . . . . . . . . 61G. Sreenivasulu, P. V. S. Srinivas and A. Goverdhan

Ontology-Driven Framework to Automatically Generate UniversityTimetable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77Shruti Bajpai and Amrita Chaturvedi

A Survey on Task Scheduling and Resource Allocation Methodsin Fog Based IoT Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89V. Sindhu and M. Prakash

Improved Linear Extrapolation Technique for Crop HealthMonitoring Using Hyperspectral Data . . . . . . . . . . . . . . . . . . . . . . . . . . 99Anjana Ghule and R. R. Deshmukh

Stratified Sampling-Based Data Reduction and CategorizationModel for Big Data Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107Kamlesh Kumar Pandey and Diwakar Shukla

xi

Product Sentiment Analysis Using Natural LanguageUnderstanding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123Prachi Ayare, Ranjnishkaur Sachdeo and Milind Penurkar

Software Cost Estimation for Python Projects Using GeneticAlgorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137Amrita Sharma and Neha Chaudhary

Prediction of Rise in Violence Inclined Opinions: Utility of SentimentAnalysis in the Modern World . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149Khushboo Sansanwal, Mamta Mittal and Lalit Mohan Goyal

Face Tagging and Recognition Using Inception Network and TripletLoss Generator Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161Shreya Kumari, Prachi Pandey and Maroti Deshmukh

Analyses of Factors Influencing Contraceptive Methodsand Predicting the Use of Method of Contraception . . . . . . . . . . . . . . . 175Sawant Rupali and Bakal Jagdish

Off-the-Record (OTR) Security Protocol Application in CloudEnvironment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195Leo Mrsic, Jurica Adamek and Ivan Cicek

Agricultural Marketing Transformation Through CrowdsourcingSystem for Product Price Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207Shambel Yilma Arega, Abhishek Ray and Prachet Bhuyan

Malware Detection Using Multilevel Ensemble SupervisedLearning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219Vidhi Garg and Rajesh Kumar Yadav

Image Splicing Forgery Detection Using DWT and Local BinaryPattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233Pooja Sahani Jaiprakash, Choudhary Shyam Prakash and Madhavi Desai

Discrete-Time Analysis of Communication Networks with SecondOptional Service and Negative User Arrivals . . . . . . . . . . . . . . . . . . . . . 243Rakhee and Shruti

Emoticon Prediction on Textual Data Using Stacked LSTM Model . . . . 259Mamta Mittal, Maanak Arora and Tushar Pandey

A Survey on Computational Intelligence Techniques for Internetof Things . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271J. Shreyas and S. M. Dilip Kumar

A Comparative Analysis of Over-the-Top Platforms: Amazon PrimeVideo and Netflix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283Advait Lad, Shivani Butala and Pramod Bide

xii Contents

An Empirical Analysis on Retrieval of Math Informationfrom the Scientific Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301Pankaj Dadure, Partha Pakray and Sivaji Bandyopadhyay

Imposition and Transmission of QPSK CommunicationThrough LabVIEW Using USRP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309Riya Chordia, Ronak Gupta, M. Ramya and A. Rajeswari

Reflection of IoT Ubiquitous Connectivity and Securityin Cross-industry Collaboration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323Priyanka Mekala, Supriya Goel and Swapnil Sutar

A Computationally Efficient Data-Dependent Projectionfor Dimensionality Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339Raghunadh Pasunuri and Vadlamudi China Venkaiah

Probabilistic-Based Energy-Efficient Single-Hop Clustering Techniquefor Sensor Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353Khushboo Jain, Anoop Kumar and Chandra Kumar Jha

RFID-Based Smart Shopping in IoT Environment: A Way to Becomea Smart Shopper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367Rahul Chauhan, Divya Punj and R. C. Joshi

Area Optimized High-Level Synthesis of Bubble Check Algorithmfor Check Node Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381Himanshu Sharma, Manju Choudhary, Vikas Pathak and Ila Roy

Identification and Classification of Expertise Using EyeGaze—Industrial Use Case Study with Software Engineers . . . . . . . . . . 391K. R. Chandrika, J. Amudha and Sithu D. Sudarsan

Crypto-Wills: Transferring Digital Assets by Maintaining Willson the Blockchain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407Jainam Chirag Shah, Mugdha Bhagwat, Dhiren Patel and Mauro Conti

One-Shot Digit Classification Based on Human Concept Learning . . . . 417A. K. Thushar and V. N. Manjunatha Aradhya

Genetic Algorithm Tuned Sizing Scheme for Grid Integrated HybridEnergy System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425Lakshaya Maheshwari and Hari Om Bansal

Medical Analysis Methods for Object Identification . . . . . . . . . . . . . . . . 435Ramgopal Kashyap and Surendra Rahamatkar

Low-Cost Smart Home Using Sensor Fusion and AutonomousWheel Chair: Caring Lonely Elders Specially in India . . . . . . . . . . . . . 447Shatadal Ghosh, C. Manikanta, Ashis Mondal, S. P. Pathak, M. Bagchiand Nataraj Dasgupta

Contents xiii

Optimal Reactive Power Dispatch Using Gravitational SearchAlgorithm to Solve IEEE-14 Bus System . . . . . . . . . . . . . . . . . . . . . . . . 463Indu Bala and Anupam Yadav

Comparative Analysis of Ensemble Classifier and Single BaseClassifier in Medical Disease Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . 475Samir S. Yadav, Vinod J. Kadam and Shivajirao M. Jadhav

Comparative Study on Punjabi Document Clustering . . . . . . . . . . . . . . 491Iknoor Singh, Vijay Paul Singh and Naveen Aggarwal

Appropriateness of Machine Learning Techniquesfor TCP with MANETs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503Hardik K. Molia and Amit D. Kothari

xiv Contents

Editors and Contributors

About the Editors

Dr. Jagdish Chand Bansal is an Associate Professor at South Asian University,New Delhi and a Visiting Research Fellow at Liverpool Hope University, UK.Holding a Ph.D. from the IIT Roorkee, Dr. Bansal is the Series Editor of the bookseries Algorithms for Intelligent Systems (AIS), published by Springer; Editor-in-Chief of the International Journal of Swarm Intelligence (IJSI), published byInderscience; and an Associate Editor of IEEE ACCESS, published by IEEE. He isa steering committee member and the general chair of the annual conference seriesSocProS, and general secretary of the Soft Computing Research Society (SCRS).He has published more than 60 research papers in various journals and conferenceproceedings.

Dr. Mukesh Kumar Gupta is a Professor at the SKIT, Jaipur. Holding a Ph.D. inComputer Science & Engineering from the MNIT, Jaipur, his current researchinterests include the security of web applications based on machine learningtechniques, and modeling of web applications. Dr. Gupta is a member of the IEEE& IEEE Computer Society and a Life Member of the Indian Society for TechnicalEducation (ISTE). He serves on the editorial board/reviewer board of variousprominent international conferences/journals, and has published more than 30papers in various journals and conference proceedings.

Dr. Harish Sharma is an Associate professor at the RTU, Kota. Holding a Ph.D.from the ABV - Indian Institute of Information Technology and Management,Gwalior, India, he is the current secretary and one of the founder members of theSoft Computing Research Society of India. He is an Associate Editor of theInternational Journal of Swarm Intelligence (IJSI), published by Inderscience, andhas also edited special issues of the journals Memetic Computing and Journal ofExperimental and Theoretical Artificial Intelligence. He has published more than 50papers in various journals and conference proceedings.

xv

Dr. Basant Agarwal is currently working as an Assistant Professor at the IndianInstitute of Information Technology Kota (IIIT) Kota, India. He has worked as apostdoctoral research fellow at the Norwegian University of Science andTechnology (NTNU), Norway; and as a research scientist at Temasek Laboratories,National University of Singapore (NUS). His research interests are in natural lan-guage processing, machine learning, and deep learning. He has published morethan 50 papers in various journals and conference proceedings.

Contributors

Jurica Adamek Algebra University College, Zagreb, Croatia

Naveen Aggarwal Department of Computer Science and Engineering, UniversityInstitute of Engineering and Technology, Panjab University, Chandigarh, India

Himanshu Agrawal Department of CSE, Jaypee Institute of InformationTechnology, Noida, India

J. Amudha Department of Computer Science and Engineering, Amrita School ofEngineering, Amrita Vishwa Vidyapeetham, Bengaluru, India

Shambel Yilma Arega School of Computer Engineering, Kalinga Institute ofIndustrial Technology Deemed to be University, Bhubaneswar, India

Maanak Arora Department of Computer Science and Engineering, G. B. PantGovernment Engineering College, Okhla, New Delhi, India

Prachi Ayare Department of Computer Science and Engineering, MIT School ofEngineering, ADT University, Rajbaug, Loni-Kalbhor, Pune, India

M. Bagchi Central Research and Training Laboratory, National Council ofScience Museums, Ministry of Culture, Government of India, Kolkata, India

Shruti Bajpai Indian Institute of Technology (Banaras Hindu University)Varanasi, Varanasi, India

Indu Bala Department of Applied Science, Northcap University, Gurugram,Haryana, India

Sivaji Bandyopadhyay Department of Computer Science and Engineering,National Institute of Technology Silchar, Silchar, Assam, India

Hari Om Bansal Department of Electrical and Electronics Engineering, BirlaInstitute of Technology and Science Pilani, Pilani, India

Mugdha Bhagwat Computer Engineering Department, VJTI Mumbai, Mumbai,India

Prachet Bhuyan School of Computer Engineering, Kalinga Institute of IndustrialTechnology Deemed to be University, Bhubaneswar, India

xvi Editors and Contributors

Pramod Bide Department of Computer Engineering, Sardar Patel Institute ofTechnology, Mumbai, India

Shivani Butala Department of Computer Engineering, Sardar Patel Institute ofTechnology, Mumbai, India

K. R. Chandrika ABB Corporate Research, Bengaluru, India;Department of Computer Science and Engineering, Amrita School of Engineering,Amrita Vishwa Vidyapeetham, Bengaluru, India

Amrita Chaturvedi Indian Institute of Technology (Banaras Hindu University)Varanasi, Varanasi, India

Neha Chaudhary Department of Computer Science and Engineering, ManipalUniversity, Jaipur, Rajasthan, India

Rahul Chauhan Graphic Era Hill University, Dehradun, India

Riya Chordia Department of ECE, M.B.M. Engineering College, Jodhpur,Rajasthan, India

Manju Choudhary Swami Keshvanand Institute of Technology, Jaipur, India

Ivan Cicek Algebra University College, Zagreb, Croatia

Mauro Conti Department of Mathematics, University of Padua, Padua, Italy

Pankaj Dadure Department of Computer Science and Engineering, NationalInstitute of Technology Silchar, Silchar, Assam, India

Nataraj Dasgupta Central Research and Training Laboratory, National Council ofScience Museums, Ministry of Culture, Government of India, Kolkata, India

Madhavi Desai Department of Computer Science and Engineering, R. N. G. PatelInstitute of Technology (RNGPIT), Bardoli, Gujarat, India

Maroti Deshmukh Department of Computer Science and Engineering, NationalInstitute of Technology Uttarakhand, Srinagar, Uttarakhand, India

R. R. Deshmukh Department of Computer Science and Information Technology,Dr. B.A.M. University, Aurangabad, India

Vidhi Garg Computer Engineering Department, Marwadi University, Rajkot,India

Shatadal Ghosh Central Research and Training Laboratory, National Council ofScience Museums, Ministry of Culture, Government of India, Kolkata, India

Anjana Ghule Department of Information Technology, GEC Aurangabad,Aurangabad, India

Supriya Goel C.R. Rao Advanced Institute of Mathematics, Statistics andComputer Science, Gachibowli, India

Editors and Contributors xvii

A. Goverdhan Department of CSE, Jawaharlal Nehru Technological UniversityHyderabad, Hyderabad, India

Lalit Mohan Goyal Department of Computer Engineering, J.C. Bose Universityof Science and Technology, YMCA, Faridabad, India

Ronak Gupta Department of ECE, M.B.M. Engineering College, Jodhpur,Rajasthan, India

Shivajirao M. Jadhav Dr. Babasaheb Ambedkar Technological University,Lonere, Raigad, India

Bakal Jagdish Department of Computer Science and Engieering, ShivajiraoS. Jondhale College of Engineering, Dombivli, India

Khushboo Jain Department of Computer Science and Engineering, AIM & ACT,Banasthali Vidyapeeth Tonk, Vanasthali, India

Pooja Sahani Jaiprakash Department of Computer Science and Engineering,CGPIT, Bardoli, Surat, India

Chandra Kumar Jha Department of Computer Science and Engineering, AIM &ACT, Banasthali Vidyapeeth Tonk, Vanasthali, India

R. C. Joshi Graphic Era University, Dehradun, India

Vinod J. Kadam Dr. Babasaheb Ambedkar Technological University, Lonere,Raigad, India

Ramgopal Kashyap Department of Computer Science and Engineering, AmityUniversity Chhattisgarh, Raipur, India

Amit D. Kothari Gujarat Technological University, Ahmedabad, Gujarat, India;Accenture, Bengaluru, Karnataka, India

Anoop Kumar Department of Computer Science and Engineering, AIM & ACT,Banasthali Vidyapeeth Tonk, Vanasthali, India

Arun Kumar Computer Science and Engineering, National Institute ofTechnology Rourkela, Rourkela, India

S. M. Dilip Kumar Department of Computer Science and Engineering, UniversityVisvesvaraya College of Engineering, Bengaluru, India

Shreya Kumari Department of Computer Science and Engineering, NationalInstitute of Technology Uttarakhand, Srinagar, Uttarakhand, India

Advait Lad Department of Computer Engineering, Sardar Patel Institute ofTechnology, Mumbai, India

Lakshaya Maheshwari Department of Electrical and Electronics Engineering,Birla Institute of Technology and Science Pilani, Pilani, India

xviii Editors and Contributors

C. Manikanta Central Research and Training Laboratory, National Council ofScience Museums, Ministry of Culture, Government of India, Kolkata, India

V. N. Manjunatha Aradhya Department of Computer Application, JSS Scienceand Technology University, Mysuru, India

Priyanka Mekala C.R. Rao Advanced Institute of Mathematics, Statistics andComputer Science, Gachibowli, India

Mamta Mittal Department of Computer Science and Engineering, G. B. PantGovernment Engineering College, Okhla, New Delhi, India

A. K. Mohapatra Associate Professor, Department of Information Technology,Indira Gandhi Delhi Technical University for Women (IGDTUW), New Delhi,India

Hardik K. Molia Gujarat Technological University, Ahmedabad, Gujarat, India;Government Engineering College, Rajkot, Gujarat, India

Ashis Mondal Central Research and Training Laboratory, National Council ofScience Museums, Ministry of Culture, Government of India, Kolkata, India

Leo Mrsic Algebra University College, Zagreb, Croatia

Jaison Mulerikkal Department of CSE, Jyothi Engineering College, Thrissur,Kerala, India

Mahendra Kumar Murmu Department of Computer Engineering, NationalInstitute of Technology, Kurukshetra, Kurukshetra, India

Partha Pakray Department of Computer Science and Engineering, NationalInstitute of Technology Silchar, Silchar, Assam, India

Kamlesh Kumar Pandey Department of Computer Science and Applications,Dr. HariSingh Gour Vishwavidyalaya, Sagar, Madhya Pradesh, India

Prachi Pandey Department of Computer Science and Engineering, NationalInstitute of Technology Uttarakhand, Srinagar, Uttarakhand, India

Tushar Pandey Department of Computer Science and Engineering, G. B. PantGovernment Engineering College, Okhla, New Delhi, India

Raghunadh Pasunuri School of Computer and Information Sciences, Universityof Hyderabad, Hyderabad, India

Dhiren Patel VJTI Mumbai, Mumbai, India

S. P. Pathak Central Research and Training Laboratory, National Council ofScience Museums, Ministry of Culture, Government of India, Kolkata, India

Vikas Pathak Swami Keshvanand Institute of Technology, Jaipur, India

Editors and Contributors xix

Milind Penurkar Department of Computer Science and Engineering, IndianInstitute of Information Technology Nagpur, Nagpur, Maharashtra, India

Manasa Pisipati Computer Science and Engineering, National Institute ofTechnology Rourkela, Rourkela, India

Choudhary Shyam Prakash Department of Computer Science and Engineering,Koneru Lakshmaiah Education Foundation, Vaddeswaram, Andhra Pradesh, India

M. Prakash Department of Information Technology, Karpagam College ofEngineering, Coimbatore, Tamil Nadu, India

Amrit Puhan Computer Science and Engineering, National Institute ofTechnology Rourkela, Rourkela, India

Divya Punj Graphic Era Hill University, Dehradun, India

Surendra Rahamatkar Department of Computer Science and Engineering,Amity University Chhattisgarh, Raipur, India

A. Rajeswari Department of ECE, Coimbatore Institute of Technology,Coimbatore, Tamil Nadu, India

Rakhee Department of Mathematics, Birla Institute of Technology and SciencePilani, Pilani, India

P. B. Ramkumar Department of CSE, Rajagiri School of Engineering andTechnology, Ernakulam, Kerala, India

M. Ramya Department of ECE, Coimbatore Institute of Technology, Coimbatore,Tamil Nadu, India

Abhishek Ray School of Computer Engineering, Kalinga Institute of IndustrialTechnology Deemed to be University, Bhubaneswar, India

Deepak Rohilla Department of Computer Engineering, National Institute ofTechnology, Kurukshetra, Kurukshetra, India

Ila Roy Swami Keshvanand Institute of Technology, Jaipur, India

Sawant Rupali Department of Information Technology, Sardar Patel Institute ofTechnology, Mumbai, India

Ranjnishkaur Sachdeo Department of Computer Science and Engineering,Indian Institute of Information Technology Nagpur, Nagpur, Maharashtra, India

Khushboo Sansanwal G. B. Pant Government Engineering College, Okhla, NewDelhi, India

Vijay Bhaskar Semwal Computer Science and Engineering, National Institute ofTechnology Bhopal, Bhopal, Madhya Pradesh, India

xx Editors and Contributors

Jainam Chirag Shah Computer Engineering Department, VJTI Mumbai,Mumbai, India

Amrita Sharma Department of Computer Science and Engineering, ManipalUniversity, Jaipur, Rajasthan, India

Himanshu Sharma Swami Keshvanand Institute of Technology, Jaipur, India

J. Shreyas Department of Computer Science and Engineering, UniversityVisvesvaraya College of Engineering, Bengaluru, India

Shruti Department of Mathematics, Birla Institute of Technology and SciencePilani, Pilani, India

Diwakar Shukla Department of Computer Science and Applications,Dr. HariSingh Gour Vishwavidyalaya, Sagar, Madhya Pradesh, India

V. Sindhu Department of Information Technology, Karpagam College ofEngineering, Coimbatore, Tamil Nadu, India

Iknoor Singh Department of Computer Science and Engineering, UniversityInstitute of Engineering and Technology, Panjab University, Chandigarh, India

Vijay Paul Singh Department of Computer Science and Engineering, UniversityInstitute of Engineering and Technology, Panjab University, Chandigarh, India

G. Sreenivasulu Department of CSE, J B Institute of Engineering andTechnology, Hyderabad, India

P. V. S. Srinivas Department of CSE, Sreenidhi Institute of Technology andScience, Hyderabad, India

Sithu D. Sudarsan ABB Corporate Research, Bengaluru, India

Mani V. Sumini Department of CSE, Rajagiri School of Engineering andTechnology, Ernakulam, Kerala, India

Swapnil Sutar C.R. Rao Advanced Institute of Mathematics, Statistics andComputer Science, Gachibowli, India

Paulsy Tharakan Department of CSE, Jyothi Engineering College, Thrissur,Kerala, India

A. K. Thushar Department of Computer Science Engineering, Jain University,Bengaluru, India

Vadlamudi China Venkaiah School of Computer and Information Sciences,University of Hyderabad, Hyderabad, India;C.R. Rao Advanced Institute of Mathematics, Statistics and Computer Science(AIMSCS), University of Hyderabad Campus, Central University Post Office,Hyderabad, Telangana, India

Editors and Contributors xxi

Anupam Yadav Department of Mathematics, Dr. B. R. Ambedkar NationalInstitute of Technology Jalandhar, Punjab, India

Priyanka Yadav IGDTUW, New Delhi, India

Rajesh Kumar Yadav Computer Engineering Department, Marwadi University,Rajkot, India

Samir S. Yadav Dr. Babasaheb Ambedkar Technological University, Lonere,Raigad, India

xxii Editors and Contributors

Extremely High-Capacity, Secure ImageSteganography Based on CompositeEdge Detection, with Message Integrity

Priyanka Yadav and A. K. Mohapatra

Abstract Image steganography is a technique for obscuring secret message withinimage pixel matrix. It provides means for secretly communicating between twoentitieswith the third party totally unaware of the embeddedmessage.One of themostimportant methods for image steganography is edge detection technique, whereinsecret data is inserted in LSBs of image edge pixels. Data embedding in edge pixelsdoes not create notable distortion in image when compared to embedding data innon-edge pixels. Different types of edge pixel detectors are available including mostrenowned Canny edge detector, Sobel edge detector, Prewitt edge detector and manyothers. Hybrid edge detection uses a combination of different edge detectors forgenerating edge pixels of image. Message to be secretly communicated is insertedin the LSB of these edge pixels, and the image thus formed called stego image issent to the concerned party. Major drawback of hybrid edge detection technique isits capacity. The number of edge pixels is very less in any image. In the last decade,many papers have focused on increasing capacity of edge detection by using differenttechniques. In this work, we have proposed an extremely high-capacity hybrid edgedetection technique by using data compression. Also, data is encrypted and theninserted in the cover image pixels. For message integrity, SHA512 hash function isused. Subsequently, this work not only improves capacity of edge detection whilemaintaining the image quality but also provides security and message integrity.

Keywords Steganography · Encryption · SHA512 · Compression · Edgedetection · Canny · Sobel · Prewitt

P. Yadav (B)IGDTUW, New Delhi, Indiae-mail: priyayadav2511@gmail.com

A. K. MohapatraAssociate Professor, Department of Information Technology, Indira Gandhi Delhi TechnicalUniversity for Women (IGDTUW), New Delhi, Indiae-mail: mohapatra.amar@gmail.com

© Springer Nature Singapore Pte Ltd. 2020J. C. Bansal et al. (eds.), Communication and Intelligent Systems, Lecture Notesin Networks and Systems 120, https://doi.org/10.1007/978-981-15-3325-9_1

1

2 P. Yadav and A. K. Mohapatra

1 Introduction

With proliferation in Internet in the last decade, data services have become veryeconomical and easy to use. With advent of Android applications, services are easilyaccessible to millions of users. These days, social networking has gained tremendouspopularity. Social networking applications can be used as and when required quiteeasily. Sharing of images and other data can be done in microseconds. With largeexchange of media in social networking applications, there comes requirement forsecure exchange of data between two users. There are two methodologies for securedata exchange between two parties: cryptography and steganography.

Steganography is an alternative approach to cryptography. Steganography offersan extra level of security in the sense that third party is unaware of the presence ofany message in the image. It makes use of the fact that human eyes are not able todistinguish very small change in image due to message embedded in LSBs of imagepixels.Andhence, third party is totally unaware of anykind of secretmessage sharing,whereas in cryptography everybody is mindful that mystery communication is goingon between included parties. Steganography is done in two diverse areas: spatialdomainmethods and frequency domain (includes different types of transformations),and each one comes with its own pros and cons [1]. Least significant bit (LSB)substitution, edge detection and pixel value difference (PVD) are some prominentspatial domain techniques used for image steganography.

Figure 1 shows basic image steganography process. At the sender side, we have“cover image” and “message”. Confidential message is inserted in the cover imagepixels without notable deformation in the image. The small amount of variationwhich is caused in image is not visible directly to human eyes due to limitationof human visual system. The resultant image so formed after inserting message in

Steganography is the practice of concealing messages or information within other non -secret text or data.

Steganography is the practice of concealing messages or information within other non secret text or data.

EMBEDDING ALGORITHM

EXTRACTING ALGORITHM

COVER IMAGE

SECRET MESSAGE

STEGO IMAGE STEGO IMAGE

RECOVERED MESSAGE

COMMUNICATION CHANNEL

KEY

KEY

Fig. 1 Basic image of steganography

Extremely High-Capacity, Secure Image Steganography … 3

cover image is called “stego image”. This stego image is transmitted to receiver viacommunication channel. Receiver extracts the message from the stego image usingsame algorithm as was used for inserting the message at transmitter end. Thus, thereceiver is able to comprehend the secret message send by the sender with third partytotally unaware of the message which is secretly inserted in cover image. This is howthe entire process of insertion and extraction of secret message is done.

In the present years, different steganography techniques are employed in differentdigital forms (image, audio, video) and networking protocols for achieving secretinformation sharing between two parties. Some major applications where imagesteganography has been used are: authentication of images in social media platforms[2, 3], for confidentiality of data stored in cloud and for its secret sharing and authen-tication purposes, for secretly sending data between user and cloud, and to securenode localization in wireless sensor networks (WSNs) [4].

Image steganography is characterized by threemain properties: embedding capac-ity, image quality (PSNR) and security. All three cannot be achieved simultaneously.If embedding capacity is increased, then image quality will decrease. And for achiev-ing security, there is a compromise on embedding capacity and algorithm complexity.A good image steganography technique is onewhich provides increased capacity andsecurity maintaining the required level of image quality.

Edge detection-based image steganography is one of the prominent methods,wherein firstly edge pixels of cover image are found using any of the edge detec-tion mechanisms or a combination of different edge detection mechanisms. Edgepixels are areas in image where sharp change in intensity of image occurs. Insert-ing message at these discontinuous locations in image helps to conceal messagefrom direct human eye detection due to limitation of human visual system. Thesecret message is then inserted in these edge pixels of the cover image to form stegoimage. At receiver end, message embedded in the stego image is extracted fromthe edge pixels. Receiver must have information about edge pixels within the stegoimage itself. Edge-based image steganography can be made more efficient in termsof capacity by using some compression technique and can be made secure by usingencryption mechanisms. Secret message should be compressed and encrypted toform “processed message” before embedding in image pixels, and at the receiverside the corresponding decryption and decompression must be done after extractingthe “processed message”.

2 Literature Review

In the last decade, many research papers in field of image steganography (spatialdomain) using edge detection have been published. One of the major challenges inedge detection-based image steganography is that the number of edge pixels is veryless; hence, capacity of such technique is limited. Also, edge pixel information mustbe available to the receiver to allow extracting of information from the edge pixels ofstego image; hence, extra information about edge pixels of image is to be inserted in

4 P. Yadav and A. K. Mohapatra

the same stego image. Consequently, it is challenging to increase capacity since secretmessage as well as edge pixel information must be embedded in same stego image.Different techniques can be employed to increase capacity of edge detection-basedimage steganography system. Many research papers have presented huge number oftechniques for increased capacity by using different methodologies.

In [5], a novel steganography technique is presented which offers increase incapacity while maintaining image imperceptibility and robustness. It offers capacityof 1.236 bpp (ratio of number of bits embedded in message to the total numberof pixels available in cover image) maintaining PSNR of 43.6209 dB and MSE of2.8248 for 512 × 512 colour image (Mandrill.tiff and Airplane.tiff) from standardimage database. The method proposed in this uses Canny edge method for detectionof edge areas in a modified form of cover image followed by applying dilationoperator. Data is inserted in some of the LSB planes of modified image using XORcoding technique which provides increased capacity along with security. However,the embedding capacity in this method still seems very low which could have beenincreased by using some compression technique in the input message.

In [6], an information hiding technique is presented which provides high capacity,message encryption (RC4) and early verification ofmessage tampering using embed-ded logo detection. It uses combination of Canny and Prewitt edge detectionmethodsfor forming hybrid edge detector. Data is inserted in blue and green planes of colourimage, and red plane carries the information about edge and non-edge image pixels.For image size of 512 × 512 for image Lena, it provides PSNR of 57.8723 dB forembedding capacity of 38,427. For image size of 512 × 512 for image Peppers, itprovides PSNR of 58.0023 dB for embedding capacity of 35,946. Also, comparingthese results with results from other papers shows that there is increase in capacitycompared to other proposed methods. Although the method proposed offers datasecurity by using RC4 encryption, the capacity of proposed method still seems lowand the image quality does not seem to be good considering such low embeddingcapacity.

In [7], a high payload image steganography technique is presented based on hybridedge detector which comprises of fuzzy and Sobel/Laplacian which provides largernumber of edge pixels. It presented bit capacity of 30,987 bits with PSNR= 45.12 dBfor 128 × 128 image size for Lena which was an improvement in bit capacity whencompared to the previous paper of [8].

In [9], cover image interpolation technique is used to increase embedding capacityalong with conventional edge detection method with no extra information to beembedded for differentiating between edge and non-edge pixels of image. For Lenaimage, it provides bit capacity of 389,976 bits with PSNR at 44.88 dB.

In [10], embedding capacity is increased by using adaptive number of LSB substi-tution method based on limitation that human eyes cannot differentiate the change inintensity. Maximum 24 bits of information are required to ensure that this adaptabil-ity is easily extractable at receiver end from stego image. For standard Lena imageof 512 × 512, it provides PSNR = 40.22 for embedding capacity of 618,160 bits.

In [11], Huffman encoding is used to compress the secret message, 2K correctiontechnique is used to decrease the difference between actual image and stego image,

Extremely High-Capacity, Secure Image Steganography … 5

andCanny edge detector is used for determining edge pixels. It is applied on greyscaleimage of Lena (512× 512) which provides embedding capacity of 38,427 bits whilemaintaining PSNR at 63.4807 dB. For Peppers greyscale image of 512 × 512, itprovides embedding capacity of 35,946 bits with PSNR at 63.2399 dB.

In [12], a novel technique for information hiding is presented which providesmore embedding capacity since edge pixel information is not needed to be stored instego image. Edge pixels of image are generated by modified cover image which isobtained by replacing 5LSBs of each pixel in image by bit ‘0’. And at the receiver endalso for generating edge pixels, the 5 LSBs of the stego image are replaced by bit ‘0’.In this way, edge pixel information generated is same for the actual image and stegoimage and no extra edge information is to be inserted in cover image and thus thisprovides extra space for embedding payload bits. For Lena (128 × 128), it providesPSNR = 30.095 dB for capacity of 67,355 using only Canny edge detection methodfor determining edge areas in image. Although this method provides advantage ofsaving space for sending edge pixel bits, 5 LSBs are used for embedding information;hence, image quality is quite compromised.

In [13], for greyscale image of Lena (512× 512), it provides embedding capacityof 771233 bits with PSNR at 38.98 dB whereas for Peppers (512 × 512) it providescapacity of 787,880 bits with PSNR at 38.27 dB. It provides increase in capacityby detecting the edge areas after clearing 5 LSBs of all pixels of actual image, thusavoiding the addition of edge pixel information to be sent in stego image. It alsoprovides 2K error correction for better imperceptibility.

In [3], both edge and non-edge pixels are used for embedding secret datawhich canhelp in authentication of parties involved in interaction in any social media platform.This authentication information is embedded in cover image and sent to concernedparty. In this way, image steganography can be used for authentication of digitalimages.

In [4], it is shown how steganography can be used to securely detect locationof sensor nodes in a wireless sensor network which is more susceptible to securitybreach in non-secure locations which can be intruded by enemies. A node sends arequest to cluster head consisting of its node ID and a cover image. Cluster head inreturnwill provide its position by embedding it secretly in the cover image. This stegoimage when received by node is used to extract its exact location. This technique issecure in the sense that only the node having that cover image will be able to get itsposition in the network while any other malicious node which tries to impersonatethe actual node will never be able to become part of authentic network.

In [14], the input message is converted into a different base number system whichprovides security and then it is embedded in cover image using some specific modulooperations which help in increasing capacity by reducing the number of bits to beadded in cover image.

In [15], a simple Canny edge detector is used for detecting change in intensity incover image and for security of data embedded OTP scheme is used. Although OTPscheme offers very high security, it requires the key size to be same or larger thanthat of input data; hence, use of OTP makes it practically unusable for applicationswhere size of input message is extremely large.

6 P. Yadav and A. K. Mohapatra

In the comprehensive literature review, we have seen how different techniqueshave been employed for increasing capacity, security and image quality of stegoimage. Also, we have seen a different variety of applicationswhich can employ imagesteganography in different domains. In the next section, we present our proposedmodel.

3 Proposed Method

In our proposed method for image steganography, many additional features are pro-vided along with basic steganography. Hybrid edge detection mechanism is used fordetection of edge as well as non-edge pixels in cover image. In order to increasecapacity, proposed mechanism uses a compression function to compress the actualsecret message into lesser number of bits. Proposed mechanism also provides secu-rity of compressed secret message by encrypting the compressedmessagewith secretencryption key. Furthermore, integrity check of actual secretmessage is also providedusing SHA512 hash methodology.

Proposedmechanism broadly consists of the following different functions: hybridedge detection, compression, encryption and SHA512 hash. Hybrid edge detectionmechanism uses a combination of three different edge detectors, namely: Canny,Sobel and Prewitt edge detectors. RGB cover image is first split into three differentimage planes, namely: red, green and blue image planes. Green and blue planes aresubjected to each edge detector to find the edge pixel matrix. Final edge pixel matrixis obtained by taking logical OR of output of all three edge detectors. Green andblue image planes are used to embed secret message, whereas red plane is used forcarrying the edge as well as non-edge pixel information for green and blue planes.Red plane is also used for embedding 512 hash codes for the actual secret message.Before the secret message is inserted in edge pixels of image planes, the message iscompressed using a compression function. This compression stage is themost crucialstage of our proposed algorithm because it is this stage which helps in increasing thebit capacity of our proposed algorithm by many times the capacity of cover image.Compression is higher and better if the secret message length is large, for very smallmessage length compression may yield a compressed output which is even largerthan size of the actual input message. This is the reason our proposed algorithmworks well for large and very large input message lengths. For providing security tothe embedded message, it is important that some kind of encryption scheme is usedbefore embedding the message in edge pixels of cover image. Here in our proposedscheme, we use AES encryption function for encrypting our compressed message.AES function used generates an encrypted output whose length is different from thecompressed input. Output length is not fixed and is variable. This variable lengthoutput further provides security. For providing integrity, the hash code for the actualsecret input message is calculated using SHA512 hash function and is embeddedin the LSBs of red image plane. This hash code helps in verifying the receiver thatcorrect message has been received at its end and the message has not been modified

Extremely High-Capacity, Secure Image Steganography … 7

by any intruder. Hence, our proposed mechanism provides integrity check usingSHA512 hash function.

At sender end, “embedding algorithm” is used which generates stego image usingthe cover image and secret message. At receiver end, “extracting algorithm” extractsthe secret message from the stego image. Steps involved in embedding and extractingalgorithm are explained in detail in following sections:

Embedding Algorithm

Figure 2 shows pictorial explanation for embedding algorithm steps. Steps forembedding algorithm (along with supportive equations) are:

1. Cover image processing: RGB cover image C[RGB] is split into correspondingred (R), green (G) and blue (B) planes. Green and blue image planes are used forembedding processed message.

SPLIT C[RGB] = [R], [G], [B]

PROCESSED MESSAGE

AES KEY

COMPRESSION AES ENCRYPTION

SHA512 EMBEDD(LSB OF RED)

COVER IMAGE

EMBEDD(EDGE PIXELS OF G,B)

INPUT MSG

INPUT MSG

COMPRESSION EMBEDD(LSB OF RED)

HYBRID EDGE

MATRIX(G(E),B(E))

STEGO IMAGE

COVER IMAGE

HYBRID EDGE DETECTION

HYBRID EDGE

MATRIX(G(E),B(E))

STEGO IMAGE(RED)

Fig. 2 Embedding algorithm

8 P. Yadav and A. K. Mohapatra

2. Hybrid Edge Detection of Green and Blue Image Planes: Hybrid edge detectorused in our proposed method consists of combination of three different edgedetectors, namely: Canny edge detector, Prewitt edge detector and Sobel edgedetector. Output of all these edge detectors is logically O red to obtain final edgepixel matrix. Figure 3 shows hybrid edge detector. Edge pixel matrix obtainedin this step for both green and blue image planes is to be embedded in LSBs ofred image plane. This will provide edge pixel information to the receiver. Step 5shows embedding procedure for these edge pixel matrixes in red image plane.

[G] → Canny = [GC][G] → Sobel = [GS][G] → Prewitt = [GP][G(E)] = [GC] OR [GS] OR [GP][B] → Canny = [BC][B] → Sobel = [BS][B] → Prewitt = [BP][B(E)] = [BC] OR [BS] OR [BP]

3. Input Message Processing: Input message (M) is firstly compressed to a smallerlength by using Zlib compression algorithm in Java. Compressedmessage is then

GREEN IMAGE PLANE

PREWITT EDGE DETECTION

CANNY EDGE DETECTION

SOBEL EDGE DETECTION

LOGICAL OR

HYBRID EDGE

MATRIX(G(E))

BLUE IMAGE PLANE

PREWITT EDGE DETECTION

CANNY EDGE DETECTION

SOBEL EDGE DETECTION

LOGICAL OR

HYBRID EDGE

MATRIX(B(E))

Fig. 3 Hybrid edge detector