Intelligent Multi-Modal Data Processing

Soham Sarkar, PhD, is an Assistant Professor in the Department of Electronics and Communication Engineering of RCC Institute of Information Technology, Kolkata, India.Abhishek Basu, PhD, is an Assistant Professor and former Head of the Department of Electronics and Communication Engineering department of RCC Institute of Information Technology, Kolkata, India.Siddhartha Bhattacharyya, PhD, is a Professor of Computer Science and Engineering at CHRIST (Deemed to be University), Bangalore, India.

• List of contributors xvSeries Preface xixPreface xxiAbout the Companion Website xxv1 Introduction 1Soham Sarkar, Abhishek Basu, and Siddhartha Bhattacharyya1.1 Areas of Application for Multimodal Signal 11.1.1 Implementation of the Copyright Protection Scheme 11.1.2 Saliency Map Inspired Digital Video Watermarking 11.1.3 Saliency Map Generation Using an Intelligent Algorithm 21.1.4 Brain Tumor Detection Using Multi-Objective Optimization 21.1.5 Hyperspectral Image Classification Using CNN 21.1.6 Object Detection for Self-Driving Cars 21.1.7 Cognitive Radio 21.2 Recent Challenges 2References 32 Progressive Performance of Watermarking Using Spread Spectrum Modulation 5Arunothpol Debnath, Anirban Saha, Tirtha Sankar Das, Abhishek Basu, and Avik Chattopadhyay2.1 Introduction 52.2 Types of Watermarking Schemes 92.3 Performance Evaluation Parameters of a Digital Watermarking Scheme 102.4 Strategies for Designing the Watermarking Algorithm 112.4.1 Balance of Performance Evaluation Parameters and Choice of Mathematical Tool 112.4.2 Importance of the Key in the Algorithm 132.4.3 Spread Spectrum Watermarking 132.4.4 Choice of Sub-band 142.5 Embedding and Detection of a Watermark Using the Spread Spectrum Technique 152.5.1 General Model of Spread Spectrum Watermarking 152.5.2 Watermark Embedding 172.5.3 Watermark Extraction 182.6 Results and Discussion 182.6.1 Imperceptibility Results for Standard Test Images 202.6.2 Robustness Results for Standard Test Images 202.6.3 Imperceptibility Results for Randomly Chosen Test Images 222.6.4 Robustness Results for Randomly Chosen Test Images 222.6.5 Discussion of Security and the key 242.7 Conclusion 31References 363 Secured Digital Watermarking Technique and FPGA Implementation 41Ranit Karmakar, Zinia Haque, Tirtha Sankar Das, and Rajeev Kamal3.1 Introduction 413.1.1 Steganography 413.1.2 Cryptography 423.1.3 Difference between Steganography and Cryptography 433.1.4 Covert Channels 433.1.5 Fingerprinting 433.1.6 Digital Watermarking 433.1.6.1 Categories of Digital Watermarking 443.1.6.2 Watermarking Techniques 453.1.6.3 Characteristics of Digital Watermarking 473.1.6.4 Different Types of Watermarking Applications 483.1.6.5 Types of Signal Processing Attacks 483.1.6.6 Performance Evaluation Metrics 493.2 Summary 503.3 Literary Survey 503.4 System Implementation 513.4.1 Encoder 523.4.2 Decoder 533.4.3 Hardware Realization 533.5 Results and Discussion 553.6 Conclusion 57References 644 Intelligent Image Watermarking for Copyright Protection 69Subhrajit Sinha Roy, Abhishek Basu, and Avik Chattopadhyay4.1 Introduction 694.2 Literature Survey 724.3 Intelligent Techniques for Image Watermarking 754.3.1 Saliency Map Generation 754.3.2 Image Clustering 774.4 Proposed Methodology 784.4.1 Watermark Insertion 784.4.2 Watermark Detection 814.5 Results and Discussion 824.5.1 System Response for Watermark Insertion and Extraction 834.5.2 Quantitative Analysis of the Proposed Watermarking Scheme 854.6 Conclusion 90References 925 Video Summarization Using a Dense Captioning (DenseCap) Model 97Sourav Das, Anup Kumar Kolya, and Arindam Kundu5.1 Introduction 975.2 Literature Review 985.3 Our Approach 1015.4 Implementation 1025.5 Implementation Details 1085.6 Result 1105.7 Limitations 1275.8 Conclusions and Future Work 127References 1276 A Method of Fully Autonomous Driving in Self-Driving Cars Based on Machine Learning and Deep Learning 131Harinandan Tunga, Rounak Saha, and Samarjit Kar6.1 Introduction 1316.2 Models of Self-Driving Cars 1316.2.1 Prior Models and Concepts 1326.2.2 Concept of the Self-Driving Car 1336.2.3 Structural Mechanism 1346.2.4 Algorithm for theWorking Procedure 1346.3 Machine Learning Algorithms 1356.3.1 Decision Matrix Algorithms 1356.3.2 Regression Algorithms 1356.3.3 Pattern Recognition Algorithms 1356.3.4 Clustering Algorithms 1376.3.5 Support Vector Machines 1376.3.6 Adaptive Boosting 1386.3.7 TextonBoost 1396.3.8 Scale-Invariant Feature Transform 1406.3.9 Simultaneous Localization and Mapping 1406.3.10 Algorithmic Implementation Model 1416.4 Implementing a Neural Network in a Self-Driving Car 1426.5 Training and Testing 1426.6 Working Procedure and Corresponding Result Analysis 1436.6.1 Detection of Lanes 1436.7 Preparation-Level Decision Making 1466.8 Using the Convolutional Neural Network 1476.9 Reinforcement Learning Stage 1476.10 Hardware Used in Self-Driving Cars 1486.10.1 LIDAR 1486.10.2 Vision-Based Cameras 1496.10.3 Radar 1506.10.4 Ultrasonic Sensors 1506.10.5 Multi-Domain Controller (MDC) 1506.10.6 Wheel-Speed Sensors 1506.10.7 Graphics Processing Unit (GPU) 1516.11 Problems and Solutions for SDC 1516.11.1 Sensor Disjoining 1516.11.2 Perception Call Failure 1526.11.3 Component and Sensor Failure 1526.11.4 Snow 1526.11.5 Solutions 1526.12 Future Developments in Self-Driving Cars 1536.12.1 Safer Transportation 1536.12.2 Safer Transportation Provided by the Car 1536.12.3 Eliminating Traffic Jams 1536.12.4 Fuel Efficiency and the Environment 1546.12.5 Economic Development 1546.13 Future Evolution of Autonomous Vehicles 1546.14 Conclusion 155References 1557 The Problem of Interoperability of Fusion Sensory Data from the Internet of Things 157Doaa Mohey Eldin, Aboul Ella Hassanien, and Ehab E. Hassanein7.1 Introduction 1577.2 Internet of Things 1587.2.1 Advantages of the IoT 1597.2.2 Challenges Facing Automated Adoption of Smart Sensors in the IoT 1597.3 Data Fusion for IoT Devices 1607.3.1 The Data Fusion Architecture 1607.3.2 Data Fusion Models 1617.3.3 Data Fusion Challenges 1617.4 Multi-Modal Data Fusion for IoT Devices 1617.4.1 Data Mining in Sensor Fusion 1627.4.2 Sensor Fusion Algorithms 1637.4.2.1 Central Limit Theorem 1637.4.2.2 Kalman Filter 1637.4.2.3 Bayesian Networks 1647.4.2.4 Dempster-Shafer 1647.4.2.5 Deep Learning Algorithms 1657.4.2.6 A Comparative Study of Sensor Fusion Algorithms 1687.5 A Comparative Study of Sensor Fusion Algorithms 1707.6 The Proposed Multimodal Architecture for Data Fusion 1757.7 Conclusion and Research Trends 176References 1778 Implementation of Fast, Adaptive, Optimized Blind Channel Estimation for Multimodal MIMO-OFDM Systems Using MFPA 183Shovon Nandi, Narendra Nath Pathak, and Arnab Nandi8.1 Introduction 1838.2 Literature Survey 1858.3 STBC-MIMO-OFDM Systems for Fast Blind Channel Estimation 1878.3.1 Proposed Methodology 1878.3.2 OFDM-Based MIMO 1888.3.3 STBC-OFDM Coding 1888.3.4 Signal Detection 1898.3.5 Multicarrier Modulation (MCM) 1898.3.6 Cyclic Prefix (CP) 1908.3.7 Multiple Carrier-Code Division Multiple Access (MC-CDMA) 1918.3.8 Modified Flower Pollination Algorithm (MFPA) 1928.3.9 Steps in the Modified Flower Pollination Algorithm 1928.4 Characterization of Blind Channel Estimation 1938.5 Performance Metrics and Methods 1958.5.1 Normalized Mean Square Error (NMSE) 1958.5.2 Mean Square Error (MSE) 1968.6 Results and Discussion 1968.7 Relative Study of Performance Parameters 1988.8 Future Work 201References 2019 Spectrum Sensing for Cognitive Radio Using a Filter Bank Approach 205Srijibendu Bagchi and Jawad Yaseen Siddiqui9.1 Introduction 2059.1.1 Dynamic Exclusive Use Model 2069.1.2 Open Sharing Model 2069.1.3 Hierarchical Access Model 2069.2 Cognitive Radio 2079.3 Some Applications of Cognitive Radio 2089.4 Cognitive Spectrum Access Models 2099.5 Functions of Cognitive Radio 2109.6 Cognitive Cycle 2119.7 Spectrum Sensing and Related Issues 2119.8 Spectrum Sensing Techniques 2139.9 Spectrum Sensing in Wireless Standards 2169.10 Proposed Detection Technique 2189.11 Numerical Results 2219.12 Discussion 2229.13 Conclusion 223References 22310 Singularity Expansion Method in Radar Multimodal Signal Processing and Antenna Characterization 231Nandan Bhattacharyya and Jawad Y. Siddiqui10.1 Introduction 23110.2 Singularities in Radar Echo Signals 23210.3 Extraction of Natural Frequencies 23310.3.1 Cauchy Method 23310.3.2 Matrix Pencil Method 23310.4 SEM for Target Identification in Radar 23410.5 Case Studies 23610.5.1 Singularity Extraction from the Scattering Response of a Circular Loop 23610.5.2 Singularity Extraction from the Scattering Response of a Sphere 23710.5.3 Singularity Extraction from the Response of a Disc 23810.5.4 Result Comparison with Existing Work 23910.6 Singularity Expansion Method in Antennas 23910.6.1 Use of SEM in UWB Antenna Characterization 24010.6.2 SEM for Determining Printed Circuit Antenna Propagation Characteristics 24110.6.3 Method of Extracting the Physical Poles from Antenna Responses 24110.6.3.1 Optimal Time Window for Physical Pole Extraction 24110.6.3.2 Discarding Low-Energy Singularities 24210.6.3.3 Robustness to Signal-to-Noise Ratio (SNR) 24310.7 Other Applications 24310.8 Conclusion 243References 24311 Conclusion 249Soham Sarkar, Abhishek Basu, and Siddhartha BhattacharyyaReferences 250Index 253