Intelligent Technologies and Applications: Third International Conference, INTAP 2020, Gjøvik, Norway, September 28–30, 2020, Revised Selected Papers ... in Computer and Information Science)

This book constitutes the refereed post-conference proceedings of the Third International Conference on Intelligent Technologies and Applications, INTAP 2020, held in Grimstad, Norway, in September 2020. The 30 revised full papers and 4 revised short papers presented were carefully reviewed and selected from 117 submissions. The papers of this volume are organized in topical sections on image, video processing and analysis; security and IoT; health and AI; deep learning; biometrics; intelligent environments; intrusion and malware detection; and AIRLEAs. Preface Organization Contents Image, Video Processing and Analysis Classification and Segmentation Models for Hyperspectral Imaging - An Overview 1 Introduction 2 Methodology 2.1 Semi-supervised Technique Used for Hyperspectral Imaging Using Active Learning and Multinomial Logistic Regression 2.2 Classification Based on Generalized Composite Kernels Framework 2.3 Spectral-Spatial Classification Using LBP 2.4 Multiple Feature Learning of HI Classification 2.5 Semi-supervised GCKs 2.6 Alpha-Expansion Algorithm 2.7 Hyperspectral Imaging-Based Dataset 3 Results and Discussion 3.1 Produced Results Through Proposed Method 4 Conclusions References Exploring Circular Hough Transforms for Detecting Hand Feature Points in Noisy Images from Ghost-Circle Patterns 1 Introduction 2 Related Work 3 Method 3.1 Assumptions 3.2 Preprocessing 3.3 Circle Detection 3.4 Overlapping Circle Detection 3.5 Circle Classification 4 Results and Discussions 5 Conclusions References Occupancy Flow Control - Case Study: Elevator Cabin 1 Introduction 2 Related Work 3 Proposed Methodology 3.1 Occupancy Analysis Inside the Elevator Cabin 3.2 Occupancy Analysis Outside the Elevator Door 3.3 Data Acquisition from the Elevator Controller 3.4 Data Fusion for Occupancy Analysis Inside the Elevator Cabin 4 Experimental Results 5 Conclusion References Security and IoT Protecting IoT Devices with Software-Defined Networks 1 Introduction 2 Attacks on IoT Devices 2.1 Access Attacks 2.2 Availability Attacks 2.3 GNSS Spoofing Attacks 2.4 Man-in-the-Middle Attacks 2.5 Masquerade Attacks 2.6 Node Attacks 3 Software-Defined Network Based Protection Methods 3.1 Access Attacks Protection 3.2 Availability Attacks Protection 3.3 GNSS Spoofing Attacks 3.4 Man-in-the-Middle Attacks 3.5 Masquerade Attacks 3.6 Node Attacks 4 Evaluation of Protection Methods 4.1 Use Case Protection Application 4.2 Access Attacks Protection Evaluation 4.3 Availability Attacks Protection Evaluation 5 Conclusion References Digital Forensic Readiness in IoT - A Risk Assessment Model 1 Introduction 2 Background 2.1 Digital Forensic Potential with IoT 2.2 Evidence Identification, Collection and Preservation 2.3 Evidence Analysis and Correlation 2.4 Attack or Deficit Attribution 3 Relevant Literature 3.1 Generic Digital Forensic Framework for IoT 3.2 Forensic State Acquisition from IoT 3.3 Forensic Investigation Framework for IoT Using a Public Digital Ledger 3.4 Forensics Edge Management System 3.5 Digital Forensic Approaches for Amazon Alexa Ecosystem 4 Research Approach 5 A Risk Assessment Model for Forensic Analysis in IoT 6 Conclusion and Future Work References System Requirements of Software-Defined IoT Networks for Critical Infrastructure 1 Introduction 1.1 Used Terminology 1.2 Development Process of SDIoTN 2 SDIoTN System Requirements 2.1 Functional and Non-functional Requirements 2.2 Performance Requirements 2.3 Privacy Requirements 2.4 Quality Requirements 2.5 Security Requirements 3 SDN Features for Requirements Implementation 4 Conclusion References An Optimized IoT-Based Waste Collection and Transportation Solution: A Case Study of a Norwegian Municipality 1 Introduction 2 Literature Review 3 Current Practices and Main Features - A Case Study 3.1 Challenges and Risks 4 Waste Collection Models 4.1 Optimization Model 4.2 Prediction Model 5 Discussion and Conclusion 6 Future Recommendations A Current Technologies References Health and AI Machine Learning Based Methodology for Depressive Sentiment Analysis 1 Introduction 2 Materials and Methods 3 Results 4 Conclusion References Simultaneous Artefact-Lesion Extraction for Skin Cancer Diagnosis 1 Introduction 2 Materials and Methods 2.1 Sources and Description of Data 2.2 Methodology 3 Results 4 Discussion 5 Conclusions and Future Work References A Statistical Study to Analyze the Impact of External Weather Change on Chronic Pulmonary Infection in South Norway with Machine Learning Algorithms 1 Introduction 2 Related Work 3 Data Collection 4 Data Processing 5 Discussion and Results 6 Conclusion References Deep Learning Interpretable Option Discovery Using Deep Q-Learning and Variational Autoencoders 1 Introduction 2 Related Work 3 Background 4 Deep Variational Q-Networks 5 Experiments and Results 5.1 Experiment Test-Bed 5.2 Hyperparameters 5.3 Latent-Space Evaluation 5.4 Performance Evaluation 6 Conclusion and Future Work References Evaluating Predictive Deep Learning Models 1 Introduction 2 Related Work 3 Problem Description 3.1 Evaluating Video Predictions 3.2 A Mixed Methods Research Design 4 Research Methods 4.1 Model and Implementation 4.2 Evaluation Method and Experimental Setting 5 Results 5.1 A Visual Interpretation of Samples 5.2 The Frame-Wise Comparison 5.3 The MMRD 5.4 Comparing the MMRD to the Frame-Wise Comparison 5.5 Model Evaluation Framework 6 Conclusions and Future Work References Pre-trained CNN Based Deep Features with Hand-Crafted Features and Patient Data for Skin Lesion Classification 1 Introduction 2 Related Works 3 Methodology 3.1 Preprocessing 3.2 Segmentation 3.3 Data Augmentation 3.4 Feature Extraction 3.5 Feature Selection 3.6 Classification 4 Experimental Results 5 Conclusions References Data-Driven Machine Learning Approach for Human Action Recognition Using Skeleton and Optical Flow 1 Introduction 2 Methodology 2.1 Skeleton 2.2 Optical Flow 2.3 KNNs for Human Action Recognition 2.4 Deep Learning 2.5 Evaluation 3 Results and Discussion 3.1 Dataset 3.2 Features Extraction 3.3 Recognition Results 4 Conclusions and Future Work References Biometrics Multilingual Voice Impersonation Dataset and Evaluation 1 Introduction 2 Related Work 3 Voice Impersonation Dataset 4 Vulnerability of ASV Systems to Voice Impersonation 4.1 Training Dataset 4.2 Automatic Speaker Verification (ASV) Systems 5 Experimental Results and Discussion 5.1 Equal Error Rate (EER) Comparison 5.2 FMR vs FNMR Comparison 5.3 IAPMR Evaluation 6 Conclusion References A Survey on Unknown Presentation Attack Detection for Fingerprint 1 Introduction 2 Traditional PAD and Anomaly Detection Based PAD 2.1 Anomaly Detection 3 Known and Unknown Presentation Attack Detection for Fingerprints 3.1 Pre-processing Techniques (Software-Based) 3.2 Convolutional Neural Network (Software-Based) 3.3 Known Sensor and Known Attacks 3.4 Known Sensor and Unknown Attacks 3.5 UnKnown Sensor and Unknown Attacks 4 Conclusions and Future Directions References Hierarchical Interpolation of Imagenet Features for Cross-Dataset Presentation Attack Detection 1 Introduction 2 Related Work on Fusion and Cross-Dataset Face PAD 3 Proposed Method 3.1 Feature Extraction 3.2 Hierarchical Cosine/SLERP Interpolation 3.3 Feature Classification and Score-Fusion 4 Experimental Setup and Results 4.1 Cross-Dataset OULU Protocol 1 Dataset Train, Replay Mobile Grandtest Development/Test Set 4.2 Features and Scores 4.3 Results and Analysis 5 Conclusions and Future-Work References Cross-lingual Speaker Verification: Evaluation on X-Vector Method 1 Introduction 1.1 Related Work 2 X-Vector Based Speaker Verification System 2.1 NIST-SRE16 Trained Model 2.2 VoxCeleb Trained Model 3 Smartphone Multilingual Dataset 4 Experiments and Results 4.1 Experiment 1 4.2 Experiment 2 5 Conclusion References Fusion of Texture and Optical Flow Using Convolutional Neural Networks for Gender Classification in Videos 1 Introduction 2 Overview of Deep-Learning Architectures 3 Gender Classification Methods 4 Proposed Method 4.1 Input Data Preparation 4.2 Feature Extraction 4.3 Classification and Fusion 5 Experimental Setup and Results 5.1 Training and Testing Protocol 5.2 Results and Analysis 6 Conclusions and Future-Work References Intelligent Environments A PM 2.5 Forecasting Model Based on Air Pollution and Meteorological Conditions in Neighboring Areas 1 Introduction 2 Related Work 3 Proposed Approach 3.1 Data 3.2 Selecting Neighboring Areas 3.3 PM 2.5 Forecasting Model 4 Experimental Evaluation and Results 5 Conclusion References The Aquatic Surface Robot (AnSweR), a Lightweight, Low Cost, Multipurpose Unmanned Research Vessel 1 Introduction 2 Related Research Work 3 Mechanical Overview 3.1 Hull Design 3.2 Propulsion 4 Hardware/Software Overview 4.1 Computation and Communications 4.2 Sensors Required to Achieve All the GNC Functions 4.3 Battery Packs 4.4 Open-Source Software 4.5 Framework Architecture 5 Simulations and Experimental Results 6 Conclusions and Future Work References Intrusion and Malware Detection Empirical Analysis of Data Mining Techniques in Network Intrusion Detection Systems 1 Introduction 2 Network Intrusion Detection Systems 3 KDDCUP99 and NSL-KDD Datasets 4 Related Works 4.1 Naïve Bayes Method in Anomaly Detection 4.2 Decision Trees Method in Anomaly Detection 4.3 Support Vector Machine Method in Anomaly Detection 4.4 Artificial Neural Networks Method in Anomaly Detection 5 Evaluation Made by Intelligence Algorithms on KDDCUP99 and NSL-KDD Datasets 5.1 Preprocessing and Analysis of Various Methods on KDDCUP99 Data Set 5.2 Preprocessing and Analysis of Various Methods on NSL-KDD Data Set 6 Feature Selection 6.1 InfoGain 7 Conclusion References Deep Neural Network Based Malicious Network Activity Detection Under Adversarial Machine Learning Attacks 1 Introduction 2 Related Work 3 Preliminary Information 3.1 Adversarial Machine Learning 3.2 Adversarial Training 4 Experiments 5 Conclusion References Towards Low Cost and Smart Load Testing as a Service Using Containers 1 Introduction 2 Background and Related Work 3 Implementation and Initial Trials 4 Experiments and Results 4.1 Load Testing with Local Slaves 4.2 Load Testing with AWS EC2 Slaves 4.3 Load Test Classification Using Weka 5 Conclusions References SecurityGuard: An Automated Secure Coding Framework 1 Introduction 2 Related Works 3 The Proposed Framework 4 Conclusion References The Multi-objective Feature Selection in Android Malware Detection System 1 Introduction 2 Background 3 Methodology 3.1 Dataset 3.2 Feature Selection 3.3 Malware Detection 4 Experimental Results 4.1 Evaluation Metrics 4.2 Results 5 Conclusion References AIRLEAs Border Management Systems: How Can They Help Against Pandemics 1 Introduction 2 Related Work 2.1 Technologies and Measures for Pandemic Situations 2.2 Border Control Systems 3 Smile System Description 4 Application Scenarios 4.1 Traveller Entering a Country with a Health-Related Alert 4.2 Traveller Exiting a Country with a Health-Related Alert 4.3 Border Control to Mitigate Pandemic Entering a Country 4.4 Border Control to Mitigate Pandemic Exiting a Country 5 SMILE Examples 5.1 SMILE Traveller Application 5.2 SMILE BCPs’ System 6 Conclusions References Translating Ethical Theory into Ethical Action: An Ethic of Responsibility Approach to Value-Oriented Design 1 Introduction 2 Ethical Theories 2.1 Consequentialism and Deontology 2.2 Balancing with Intuitions 2.3 Casuistry 2.4 Value-Oriented Design Approaches 3 An Ethic of Responsibility 4 Conclusion References An Example of Privacy and Data Protection Best Practices for Biometrics Data Processing in Border Control: Lesson Learned from SMILE 1 Introduction 2 SMILE Data Governance Framework 2.1 Legal Frameworks: EU and National Laws 2.2 Entity: Organizational and Technical Measures 2.3 Data Protection Impact Assessment (DPIA) 3 Privacy and Data Protection Best Practices 4 Conclusions and Recommendations References Unsupervised Single Image Super-Resolution Using Cycle Generative Adversarial Network 1 Introduction 2 Related Works 3 Proposed Methodology 4 Experimental Analysis 4.1 Training Details and Hyper-parameter Settings 4.2 Quantitative Analysis 4.3 Qualitative Analysis 5 Conclusion References Criminal Network Community Detection in Social Media Forensics 1 Introduction 2 Background 3 Knowledge Extraction Layer 4 Results and Discussions 4.1 Data Source and Data Pre-processing 4.2 Data Graph Description 4.3 Community Detection 5 Conclusions References Data Privacy in IoT Equipped Future Smart Homes 1 Introduction 2 Smart Home 2.1 IoT Intelligent Services 2.2 Privacy Violation Scenarios 3 Data Privacy 3.1 User's Activity Pattern 3.2 Discussion 4 Conclusion and Future Work References Machine Learning and the Legal Framework for the Use of Passenger Name Record Data 1 Introduction 2 The PNR Legal Framework: Europe and Germany 2.1 Europe 2.2 Germany 3 Technological Approaches to Pattern Creation and Pattern Matching 3.1 Theory-Based Approaches 3.2 Machine Learning Approaches 4 Conclusion References Migration-Related Semantic Concepts for the Retrieval of Relevant Video Content 1 Introduction 2 Migration-Related Semantic Concepts (MRSCs) 2.1 Migration Theories 2.2 Factors Classification 3 MRSC-Based Video Retrieval 4 Experiments and Results 5 Conclusions and Future Work References Correction to: Intelligent Technologies and Applications 0 Correction to: S. Yildirim Yayilgan et al. (Eds.): Intelligent Technologies and Applications, CCIS 1382, https://doi.org/10.1007/978-3-030-71711-7 Author Index