Intelligent Computing Theories and Application: 17th International Conference, ICIC 2021, Shenzhen, China, August 12–15, 2021, Proceedings, Part I

Description

This two-volume set of LNCS 12836 and LNCS 12837 constitutes - in conjunction with the volume LNAI 12838 - the refereed proceedings of the 17th International Conference on Intelligent Computing, ICIC 2021, held in Shenzhen, China in August 2021. The 192 full papers of the three proceedings volumes were carefully reviewed and selected from 458 submissions. The ICIC theme unifies the picture of contemporary intelligent computing techniques as an integral concept that highlights the trends in advanced computational intelligence and bridges theoretical research with applications. The theme for this conference is “Advanced Intelligent Computing Methodologies and Applications.” The papers are organized in the following subsections: Evolutionary Computation and Learning, Image and signal Processing, Information Security, Neural Networks, Pattern Recognition Swarm Intelligence and Optimization, and Virtual Reality and Human-Computer Interaction. Preface Organization Contents – Part I Contents – Part II Contents – Part III Evolutionary Computation and Learning Multi-objective Optimization-Based Task Offloading and Power Control for Mobile Edge Computing 1 Introduction 2 Related Work 3 System Model 3.1 Network Model 3.2 Communication Model 3.3 Computation Model 3.4 Cost Model 4 Multi-objective Computation Offloading Algorithm Based on NSGA-II in MEC (MOPCA) 4.1 Objective Functions 4.2 The Main Steps of MOPCA 5 Simulation Study 5.1 Experiment Setup 5.2 Performance Evaluation 6 Conclusion References An Evolutionary Neuron Model with Dendritic Computation for Classification and Prediction 1 Introduction 2 Dendritic Neuron Model 2.1 Model Description 2.2 Neural Pruning Scheme 2.3 Logic Circuit Implementation 3 Learning Algorithm 3.1 Firefly Algorithm 4 Experimental Settings 5 Performance Comparison 5.1 Comparison with Other Learning Algorithms 5.2 Neural Pruning and Hardware Implementation 6 Conclusions References An Improved Genetic Algorithm for Distributed Job Shop Scheduling Problem 1 Introduction 2 IGA Algorithm 3 Computational Result 4 Discuss and Analysis 5 Conclusion References An Improved Teaching-Learning-Based Optimization for Multitask Optimization Problems 1 Introduction 2 Multitask Optimization (MTO) 3 Teaching-Learning-Based Optimization 4 Multi-tasking Teaching-Learning Based Optimization 4.1 Teaching Stage 4.2 Learning Stage 4.3 Selection Mechanism 5 Simulation Experiment 6 Conclusions References Emotion Recognition from Facial Expressions Using a Genetic Algorithm to Feature Extraction 1 Introduction 2 State of the Art 3 Proposed Method 3.1 Detection of Areas of Interest 3.2 Features Extractors 3.3 Feature Selection 3.4 Classification Techniques 4 Experimental Results 4.1 Data Sets 4.2 Feature Extraction and Selection 4.3 Parameter Selection 5 Results 6 Conclusions References An Efficient Competitive Swarm Optimizer for Solving Large-Scale Multi-objective Optimization Problems 1 Introduction 2 Preliminary 2.1 Related Works of Competitive Swarm Optimizer 3 The Proposed E-CSO 3.1 The Complete Framework of E-CSO 3.2 The Proposed Particle Updating Strategy in E-CSO 3.3 Discussions on E-CSO 3.4 Complexity Analysis 4 Experimental Studies 4.1 Benchmark Problems and Performance Metrics 4.2 The Compared Algorithms and Experimental Settings 4.3 Comparisons Between E-CSO and Existing MOEAs 5 Conclusions and Future Work References KGT: An Application Mapping Algorithm Based on Kernighan–Lin Partition and Genetic Algorithm for WK-Recursive NoC Architecture 1 Introduction and Motivation 2 Related Work 3 Problem Formulations 3.1 Energy Consumption Model 3.2 Definition of Application Mapping 4 The Proposed KGT Mapping Algorithm 5 Experimentation and Results 5.1 The Evaluation Model 5.2 Simulator and Benchmarks 5.3 Results and Analysis 6 Conclusion References Evolutionary Algorithms for Applications of Biological Networks: A Review 1 Introduction 2 Applications in Protein Complex Detection 2.1 Definition of Protein Complex Detection 2.2 GA-Based Approaches 2.3 SI-Based Approaches 3 Applications in Biological Network Alignment 3.1 Definition of Biological Network Alignment 3.2 GA-Based Approaches 3.3 SI-Based Approaches 4 Applications in Gene Regulatory Network Inference 4.1 Definition of Gene Regulatory Network Inference 4.2 GA-Based Approaches 4.3 SI-Based Approaches 5 Discussions 6 Conclusions References Computational Prediction of Protein-Protein Interactions in Plants Using Only Sequence Information 1 Introduction 2 Materials and Methodology 2.1 Data Collection and Data Set Construction 2.2 Position-Specific Scoring Matrix (PSSM) 2.3 Inverse Fast Fourier Transform 2.4 Weighted Sparse Representation-Based Classification 3 Results and Discussion 3.1 Evaluation Criteria 3.2 Assessment of Prediction Ability 3.3 Comparison of the Proposed Model with Different Classifiert 4 Conclusion and Discussion References Image and Signal Processing A Diabetic Retinopathy Classification Method Based on Novel Attention Mechanism 1 Introduction 2 Methodology 2.1 Backbone Networks Based on Transfer Learning 2.2 Pixel-Level Attention Module 2.3 Feature Fusion 2.4 Training Algorithm 3 Datasets 3.1 Data Normalization 3.2 Data Augmentation 4 Experimental and Results 4.1 Experimental Setup 4.2 Evaluation Metrics 4.3 Model Performance Analysis 5 Conclusion References A Comparable Study on Dimensionality Reduction Methods for Endmember Extraction 1 Introduction 2 The Proposed Method 3 Experimental Results 4 Conclusions References Hyperspectral Image Classification with Locally Linear Embedding, 2D Spatial Filtering, and SVM 1 Introduction 2 The Proposed Method 3 Experimental Results 4 Conclusions References A Hierarchical Retrieval Method Based on Hash Table for Audio Fingerprinting 1 Introduction 2 Audio Fingerprint Extraction Based on LLE 2.1 Pre-processing 2.2 Maximum Spectral Points Extraction 2.3 Sub-region Division 2.4 Audio Fingerprint Generation 3 Hierarchical Retrieval Based on Hash Table 3.1 Retrieval Model Overview 3.2 Database Construction 3.3 Retrieval of Query Audio 4 Experimental Results and Analysis 4.1 Experimental Data 4.2 Performance Evaluation 4.3 Results and Analysis 5 Conclusions References Automatic Extraction of Document Information Based on OCR and Image Registration Technology 1 Introduction 2 Related Work 2.1 Image Rectification and Registration 2.2 OCR 3 Text Detection and Recognition Based on Two-Branch Mask OCR Model 3.1 Framework 3.2 Label Generation 3.3 Objective Function 3.4 Forward Propagation 4 Experiments 4.1 Data of Image Rectification 4.2 Image Rectification Evaluation 4.3 Data of Text Detection and Recognition 4.4 Information Extraction and Denoising 4.5 Results 5 Conclusion References Using Simplified Slime Mould Algorithm for Wireless Sensor Network Coverage Problem 1 Introduction 2 Wireless Sensor Network Coverage Optimization 3 Slime Mould Algorithm 3.1 Mathematical Model 4 Simplified Slime Mould Algorithm 5 Simulation Result 5.1 Problem Definition 5.2 Comparative Analysis of Experiments 5.3 Analysis of Statistical Significance 6 Conclusions References Super-Large Medical Image Storage and Display Technology Based on Concentrated Points of Interest 1 Introduction 2 Principles and Methods 2.1 Problem Description 2.2 Image Preprocessing 2.3 Image Stitching and Cutting Based on Spatial Coordinates 2.4 Loading Mechanism Based on Pyramid Data Structure 2.5 Quick Display Mechanism Based on Concentrated Points of Interest 3 Experimental Realization 4 Conclusion References Person Re-identification Based on Hash 1 Introduction 2 Related Work 3 Method 3.1 Overall network structure 3.2 Deep Feature Learning 3.3 Hash Mapping Function Learning 3.4 Search Strategies from Coarse to Fine 4 Experiment 4.1 Datasets and Evaluation Indicators 4.2 Experimental Environment and Parameter Setting 4.3 Experimental Results and Analysis 5 Conclusion References A Robust and Automatic Recognition Method of Pointer Instruments in Power System 1 Introduction 2 Related Works 3 Automatic Instrument Reading Method 3.1 Instrument Positioning 3.2 Image Preprocessing 3.3 Pointer Reading Recognition Based on DSF 4 Experiment 4.1 Experiment on the Influence of Illumination Factors 4.2 Error Discussion 5 Conclusion References Partial Distillation of Deep Feature for Unsupervised Image Anomaly Detection and Segmentation 1 Introduction 2 Related Work 2.1 Image-Reconstruction Based Methods 2.2 Feature-Embedding Based Methods 3 Method 3.1 Partial Knowledge Distillation 3.2 Anomaly Scoring 4 Experiments 4.1 Experimental Setup 4.2 Anomaly Detection 4.3 Anomaly Segmentation 4.4 Ablation Study 5 Conclusion References Speech Recognition Method for Home Service Robots Based on CLSTM-HMM Hybrid Acoustic Model 1 Introduction 2 Hybrid Speech Features TEOGFCC + ∆TEOGFCC Extraction Method in Domestic Environment 2.1 Related Basic Theories 2.2 Hybrid Speech Features TEOGFCC + ∆TEOGFCC Extraction Based on GFCC and TEO 3 Speech Recognition Based on CLSTM-HMM 3.1 Network Structure of the CLSTM-HMM Hybrid Speech Acoustic Model 3.2 Training of the CLSTM-HMM Hybrid Acoustic Model 3.3 Comparison of Acoustic Model Results 4 Experiment of Home Service Robots 4.1 Experimental Results and Analysis 4.2 Simulation Results 5 Conclusion References Serialized Local Feature Representation Learning for Infrared-Visible Person Re-identification 1 Introduction 2 Related Works 3 Proposed Method 3.1 Dual-Path Structure 3.2 Serialized Local Feature Representation Learning 4 Experiments 4.1 Dataset Description 4.2 Implementation Details 4.3 Comparison with Other Methods 5 Conclusions References A Novel Decision Mechanism for Image Edge Detection 1 Introduction 2 Proposed Method 2.1 The Imaginary Part of Multi-directional Gabor Filters 2.2 Confidence Interval 2.3 New Edge Detection Measure 3 Experiment Results and Performance Evaluation 3.1 The Evaluation of ROC 3.2 FOM Evaluation 4 Conclusions References Rapid Earthquake Assessment from Satellite Imagery Using RPN and Yolo v3 1 Introduction 2 Proposed Methodology 2.1 Dataset 2.2 Preprocessing 2.3 Region Proposal Network 2.4 Region of Interest 2.5 Yolo V3 2.6 Building Extraction 2.7 Assessment 3 Results and Analysis 3.1 Morphological Operators 3.2 Mask R-CNN 3.3 Yolo V3 3.4 Combined Approach 3.5 Custom Dataset Assessment 3.6 Evaluations 4 Conclusion References Attention-Based Deep Multi-scale Network for Plant Leaf Recognition 1 Introduction 2 Related Work 2.1 Res2Net 2.2 ECANet 3 DMSNet (Deep Multi-scale Network) 4 ADMSNet (Attention-Based Deep Multi-scale Network) 5 Experiment 5.1 Experimental Environment and Parameter Setting 5.2 Experimental Dataset 5.3 Classification Experiments with Real Data 5.4 Ablation Experiments of ADMSNet 6 Conclusions References Information Security Short Video Users’ Personal Privacy Leakage and Protection Measures 1 Introduction 2 Types of Privacy Leaks 2.1 Unconscious Disclosure 2.2 Conscious Local Protection 2.3 Consented Information Collection 3 Analysis of the Causes of Privacy Leakage in Short Videos 3.1 Optimistic Bias of Information Subjects 3.2 The Need to Build Group Identity 3.3 Privacy Agreements Obscures Users’ Wishes 4 Responses to Privacy Protection in Short Videos 4.1 Raising Awareness of Information Security Among Short Video Users 4.2 Short Video Platform Governance 4.3 Short Video Industry Self-discipline 5 Conclusion References An Efficient Video Steganography Method Based on HEVC 1 Introduction 2 Theoretical Framework 2.1 Intra-frame Prediction 2.2 Intra-frame Distortion Drift 2.3 Matrix Encoding 3 Description of Algorithm Process 3.1 Embedding 3.2 Data Extraction 4 Case Study 5 Conclusion References Analysis on the Application of Blockchain Technology in Ideological and Political Education in Universities 1 Introduction 2 Background of Blockchain Based Ideological and Political Education 3 Proposed Scheme of Blockchain Based Ideological and Political Education 3.1 Elements 3.2 Chain Resource Network Assumption 4 Challenges and Solutions to the Implementation of the Proposed Scheme in Ideological and Political Education 4.1 Construction of Ideological and Political Education Database Server in Colleges and Universities 4.2 Enthusiasm of Ideological and Political Educators in Universities 4.3 College Students’ Initiative and Consciousness in Learning Resources 5 Conclusion References Parallel Security Video Streaming in Cloud Server Environment 1 Introduction 2 Related Works 3 Real-Time Characteristics of Video Security Services 4 Performance Evaluation of PSS Service 5 Conclusions References An Efficient Video Steganography Scheme for Data Protection in H.265/HEVC 1 Introduction 2 Related Technical Backgrounds 2.1 Intrapicture Distortion Drift in H.265/HEVC 2.2 Transformation and Inverse Transformation in H.265/HEVC 3 Proposed Prediction Modes and DCT/DST Coefficients Based Video Steganography Scheme 3.1 Prediction Modes Selection for Embedding Blocks 3.2 Specific Coefficients Selection for Embedding Blocks 4 Experimental Evaluation 5 Conclusion References A Robust Lossless Steganography Method Based on H.264/AVC 1 Introduction 2 Theoretical Frame-Work 2.1 Prevention of Intra-frame Distortion Drift 2.2 BCH Code 2.3 Secret Sharing 3 Proposed Algorithm 3.1 Embedding 3.2 Extracting 4 Experiment Results 5 Conclusion References Research on Application of Blockchain Technology in Higher Education in China 1 Introduction 2 Blockchain Technology 2.1 Concept of Blockchain 2.2 Main Features of Blockchain 2.3 Types of Blockchain 3 Multi-scenario Application of Blockchain in Higher Education 3.1 Issue and Management of Certificates 3.2 Learning Record and Storage 3.3 Digital Teaching Resources Sharing 3.4 Personality Identification Solutions 4 Issues and Challenges of Blockchain in Education 4.1 The Data Storage Space Factor 4.2 The Safety Factor 4.3 The Policy and Empirical Factors 4.4 The Property Rights Disputes Factor 5 Conclusion References Neural Networks Multi-class Text Classification Model Based on Weighted Word Vector and BiLSTM-Attention Optimization 1 Introduction 2 Proposed Model 2.1 The Construction of the Weighted Word Vector 2.2 BiLSTM-Attention Module 2.3 Feedforward Neural Network Classifier 3 Experiments 3.1 Experiments Environment and Dataset 3.2 Comparison of the Performance of Weighted Word Vectors 3.3 Network Comparison 4 Conclusion References Fault Diagnosis Based on Unsupervised Neural Network in Tennessee Eastman Process 1 Introduction 2 Competitive Neural Network 2.1 The Structure of Competitive Neural Networks 2.2 Algorithm of Competitive Neural Network 3 Self-organizing Feature Mapping Neural Network 3.1 The Structure of SOM Neural Network 3.2 Algorithm of SOM Neural Network 4 Unsupervised Learning Neural Network Simulation of TEP 5 Conclusion References Constraint Interpretable Double Parallel Neural Network and Its Applications in the Petroleum Industry 1 Introduction 2 Constraint Interpretable Double Parallel Neural Network 3 Case Study 4 Conclusions References Alcoholism Detection via 5-Layer Customized Convolution Neural Network 1 Introduction 2 Dataset 3 Methodology 3.1 Convolution 3.2 Pooling 3.3 Batch Normalization 3.4 Rectified Linear Unit 3.5 Structure of Customized CNN 3.6 Measures 4 Experiment Result and Discussions 4.1 Statistical Analysis 4.2 Comparison with State-of-the-Art Approaches 5 Conclusion References A Heterogeneous 1D Convolutional Architecture for Urban Photovoltaic Estimation 1 Introduction 2 PV Estimation Problem 3 Related Work 4 1D Convolutional Neural Networks 5 Our Proposed Approach 6 Experimental Results 7 Conclusions References Adversarial Attacks and Defenses in Deep Learning: A Survey 1 Introduction 2 Adversarial Attacks 2.1 L-BFGS 2.2 Attack on Gradient 2.3 Carlini and Wagner Attack (C&W Attack) 2.4 Deepfool 2.5 GAN-Based Attacks 2.6 Universal Adversarial Perturbations 2.7 Transfer-Based Attacks 3 Adversarial Defense 3.1 Reduce or Eliminate the Adversarial Attack 3.2 Enhance the Robustness of Network 3.3 Ensemble Learning 3.4 Detect Only 4 Discussions 4.1 Attack and Defense 4.2 Defense and Model Performance 5 Conclusions References Pattern Recognition Fine-Grained Recognition of Crop Pests Based on Capsule Network with Attention Mechanism 1 Introduction 2 Related Works 2.1 Attention Mechanism 2.2 Capsule Network (CNet) 3 Capsule Network with Attention Mechanism 4 Experiments and Results 5 Conclusions References Small Object Recognition Based on the Generative Adversarial Network and Multi-instance Learning 1 Introduction 2 Small Object Recognition Based on GAN and CNN 2.1 Framework 2.2 Data Augmentation Based on GAN 2.3 Small Object Recognition Based on MIL 3 Experimental Results and Analysis 4 Conclusion References Deep Learning Based Semantic Page Segmentation of Document Images in Chinese and English 1 Introduction 2 Related Work 3 Methodology 3.1 Semantic Segmentation Network 3.2 Region Segmentation 4 Experiments 4.1 Datasets 4.2 Metrics 4.3 Experimental Results on Our Collected Dataset 4.4 Experimental Results on Public POD Dataset 5 Conclusion References Non-central Student-t Mixture of Student-t Processes for Robust Regression and Prediction 1 Introduction 2 Related Models and Learning Algorithms 2.1 Non-central Student-t Distribution 2.2 Gaussian Process 2.3 Student-t Process 3 The tMtP Model and Its Learning Algorithm 4 Experimental Results 4.1 Experimental Results on Synthetic Datasets. 4.2 Comparison on Synthetic Datasets 4.3 Application to a Real Word Dataset 5 Conclusion References Multi-class Tissue Classification in Colorectal Cancer with Handcrafted and Deep Features 1 Introduction 2 Materials 3 Methods 3.1 Image Features 3.2 Stain Normalization 3.3 Deep Learning Models 4 Experimental Results 4.1 Discussion and Explainability 5 Conclusions and Future Works References Plant Leaf Recognition Network Based on Fine-Grained Visual Classification 1 Introduction 2 DataSet and Methods 2.1 DataSet 2.2 Method 2.3 Loss Function 3 Architecture 4 Experiment 4.1 Implementation Details 4.2 Result 5 Conclusion References Anomaly Detection Based on Video Prediction and Latent Space Constraints 1 Introduction 2 Related Work 2.1 Traditional Anomaly Detection Methods 2.2 Reconstruction-Based Methods 2.3 Video Prediction-Based Methods 3 Proposed Model 3.1 Network Architecture 3.2 Loss Function 3.3 Choice of Anomaly Score 3.4 Datasets 3.5 Ablation Experiment 3.6 Comparison of Experimental Results 3.7 Experimental Results Visualization and Analysis 4 Conclusions References A Robust Distance Regularized Potential Function for Level Set Image Segmentation 1 Introduction 2 Classical Distance Regularization Term 3 A Novel Logarithmic and Power Function Based Distance Regularization Potential Function 4 Experimental Results 5 Conclusion References Label Similarity Based Graph Network for Badminton Activity Recognition 1 Introduction 2 Related Work 3 Methodology 4 Experiments 4.1 Experimental Setup 4.2 Comparison with Baselines 4.3 Ablations and Analysis 5 Conclusion References MITT: Musical Instrument Timbre Transfer Based on the Multichannel Attention-Guided Mechanism 1 Introduction 2 Related Work 2.1 Generative Adversarial Networks (GANs) 2.2 Time-Frequency Representation 2.3 Image-To-Image Translation 3 Method 3.1 Spectrogram-To-Spectrogram Timbre Transfer 3.2 Multichannel Attention-Guided Mechanism 3.3 Markov Discriminator Guidance of Higher-Order Spectrogram Feature Capture 3.4 Loss Function 4 Experiments 4.1 Experimental Setup 4.2 Evaluation Mechanism 4.3 Musical Instrument Timbre Transfer Experiment 5 Summary References Classification of Benign-Malignant Pulmonary Nodules Based on Multi-view Improved Dense Network 1 Introduction 2 Dataset 3 Method 3.1 View Extraction 3.2 Multi-layer CrossDenseNet 3.3 Dense Residual Block 3.4 Fusion Algorithm 3.5 Training 4 Experiment 4.1 Exploration of Network Structural Hierarchy 4.2 Exploration of the Number Of Dense Block 4.3 Network Evaluation 5 Conclusion References Deep Convolution Neural Network Based Research on Recognition of Mine Vehicle Head and Tail 1 Introduction 2 Related Work 3 Method 3.1 Image Processing 3.2 Deep CNN 3.3 Algorithm Equations 3.4 Algorithm Flow 4 Experiment 4.1 Data Source and Composition 4.2 Parameters Setting 4.3 Comparison of Results 4.4 Compare with VGG16 and ResNet34 5 Summary and Future Work References Compact Finite-State Super Transducers for Grapheme-to-Phoneme Conversion in Highly Inflected Languages 1 Introduction 2 Language Resources 3 FST Representations of Pronunciation Dictionaries 4 Finite-State Super Transducers 4.1 Construction of FSSTs 4.2 Experimental Results 5 Non-dictionary Words 6 Discussion 7 Conclusions References Swarm Intelligence and Optimization An Improved SMA Algorithm for Solving Global Optimization Problems 1 Introduction 2 The Standard SMA Algorithm 3 The Improved Slime Mould Algorithm (iSMA) 3.1 Initialization 3.2 The Dominant-Non-dominant Swarm Partitioning Strategy 3.3 The Gauss-Cauchy Hybrid Mutation Mechanism 4 Simulation Experiment and Analysis 4.1 Benchmark Funtions 4.2 Performance Analysis and Algorithm Comparison 5 Conclusions References An Improved Chicken Swarm Optimization Algorithm with Fireworks Factor 1 Introduction 2 CSO Algorithm Introduction 3 FW-ICSO Algorithm Introduction 3.1 Elimination Mechanism 3.2 Creates New Individuals 3.3 Inertial Factor 3.4 The Flow of the FW-CSO Algorithm 4 Experimental Comparison and Analysis 4.1 Experimental Parameter Settings 4.2 Experiment Analysis 5 Conclusion and Future Work References An Improved Lagrangian Relaxation Algorithm for Solving the Lower Bound of Production Logistics 1 Introduction 2 The Vehicle Routing Problem with Time Window and Lot Size 2.1 Optimal Batch Model 2.2 The Construction of VRP Model 3 Solution Approach 3.1 Introduction of Lagrangian Relaxation Algorithm 3.2 Improved Lagrangian Relaxation Model of VRPTW 4 Test and Comparisons 4.1 Numerical Experiment 4.2 Optimal Batch Instance 4.3 Lower Bound of VRPTW 5 Conclusions and Future Research References Multidimensional Estimation of Distribution Algorithm for Distributed No-Wait Flow-Shop Scheduling Problem with Sequence-Independent Setup Times and Release Dates 1 Introduction 2 DNWFSP with SISTs and RTs 2.1 Problem Descriptions 2.2 DNWFSP with SISTs 2.3 DNWFSP with SISTs and RTs 3 TDEDA for DNWFSP with SISTs and RTs 3.1 Solution Representation 3.2 Three-Dimensional Matrix Cube 3.3 Updating Mechanism 3.4 New Population Generation 3.5 Local Search 4 Computational Result and Comparisons 5 Conclusions and Future Research References Hybrid Whale Optimization Algorithm for Solving Green Open Vehicle Routing Problem with Time Windows 1 Introduction 2 GOVRP 2.1 Problem Description 2.2 Energy Consumption Model 2.3 Mathematical Model of GOVRP 3 Algorithm Design 3.1 Population Initialization 3.2 The HWOA for GOVRP 3.3 Local Search 4 Computational Results 4.1 Settings of Parameters 4.2 Computational Results 5 Conclusions and Future Research References Hybrid Grey Wolf Optimizer for Vehicle Routing Problem with Multiple Time Windows 1 Introduction 2 Problem Description 3 HGWO for VRPMTW 3.1 Encoding and Decoding 3.2 Nonlinear Decreasing Distance Control Parameter 3.3 Variable Neighborhood Local Search 3.4 HGWO Process 4 Simulation Result and Comparisons 4.1 Experimental Setup 4.2 Results and Comparison 5 Conclusion and Future Work References Spatial Prediction of Stock Opening Price Based on Improved Whale Optimized Twin Support Vector Regression 1 Introduction 2 The Prediction Model of Stock Opening Price Range Granulated by Fuzzy Information 3 The Prediction Model of Stock Opening Price Range Using IWOA-TSVR 3.1 Twin Support Vector Regression (TSVR) 3.2 Whale Optimization Algorithm and Its Improvement 3.3 Improved Whale Optimization Algorithm 3.4 Fuzzy Information Granulation Based on IWOA-TSVR for Prediction Model of Stock Opening Price Range 4 Experimental Results and Analysis 4.1 Experimental Data and Algorithm Parameter Setting 4.2 Results and Analysis 5 Conclusions References Channel Assignment Algorithm Based on Discrete BFO for Wireless Monitoring Networks 1 Introduction 2 Related Work 3 Problem Descriptions 3.1 Network Model 3.2 Problem Formulation 4 Discrete BFO Based Channel Assignment Algorithm 4.1 Basic Theory of BFO 4.2 Process of DBFO 4.3 Update and Revise Channel Coding Table 4.4 Theoretical Performance Analysis of DBFO 5 The Experimental Results and Analysis 5.1 Simulations 5.2 Practical Network Experiment 6 Conclusion References A Multi-objective Particle Swarm Optimization Algorithm Embedded with Maximum Fitness Function for Dual-Resources Constrained Flexible Job Shop Scheduling 1 Introduction 2 Problem Description 2.1 Literature Review 2.2 Mathematical Model 3 Design of MOPSO 3.1 Maximin Fitness Function 3.2 Initialization and Maintenance of EA 3.3 Coding Scheme 3.4 Initialization of Population 3.5 Global Search Mechanism 3.6 Local Search Mechanism 3.7 Procedure of the MOPSO Algorithm 4 Simulation Results 4.1 Parameter Analysis 4.2 Comparison 5 Conclusions References An Improved Firefly Algorithm for Generalized Traveling Salesman Problem 1 Introduction 2 The Standard Firefly Algorithm 3 The Improved Firefly Algorithm 3.1 The Two-Layer Coding Method 3.2 The Inter-individual Distance Formula 3.3 The Position Update Formula 3.4 The Repair Method for Infeasible Solutions 3.5 The Improved Iterative Local Search 3.6 The C2opt Optimization Operator 3.7 The Firefly Mutation 3.8 The Improved Firefly Algorithm Flow Chart 4 Simulation Experiments and Result Analysis 4.1 The Setting of Acceptance Probability 4.2 Experimental Results and Analysis 5 Conclusion References Virtual Reality and Human-Computer Interaction User Study on an Online-Training System of Activity in Daily Life for the Visually Impaired 1 Introduction 2 Configuration of Our Online ADL Training System 3 Online ADL Training 4 User Study 5 Discussion 6 Conclusions References Person Property Estimation Based on 2D LiDAR Data Using Deep Neural Network 1 Introduction 2 Related Works 3 Proposed Method 3.1 Dataset Preparation 3.2 Person Property Estimation 4 Results and Discussion 5 Conclusion References Detection of Pointing Position by Omnidirectional Camera 1 Introduction 2 Related Works 3 Detection of Pointing Position 3.1 Overview 3.2 Dataset 3.3 Detection of Person Area 3.4 Estimation of Pointing Position 4 Experiments 4.1 Results Using Only Real Image 4.2 Results Using CG Image and Real Image 5 Conclusions References Optimization of Low-Speed Dual Rotor Axial Flux Generator Design Through Electromagnetic Modelling and Simulation 1 Introduction 2 Related Work 3 Preliminary Analysis 4 Simulation 5 Conclusion References A Lightweight Attention Fusion Module for Multi-sensor 3-D Object Detection 1 Introduction 2 Related Work 2.1 LiDAR-Based 3-D Object Detection 2.2 Multi-modal 3-D Object Detection 2.3 Attention Module 3 Proposed Method 3.1 Point Feature Fusion Module 3.2 Voxel Feature Encoder Module and 3D Backbone 4 Experiments 4.1 Dataset 4.2 Experimental Setting 4.3 Results 4.4 Ablation Study 4.5 Visualization 5 Conclusions References Regression-Aware Classification Feature for Pedestrian Detection and Tracking in Video Surveillance Systems 1 Introduction 2 Related Works 3 The Proposed Method 3.1 Cross-Global Context 3.2 Regression-Aware Classification Feature 4 Experiment Setup 5 Results 5.1 Ablation Study 5.2 Comparison with State-of-the-Art Methods 6 Conclusion References Efficient Face Detector Using Spatial Attention Module in Real-Time Application on an Edge Device 1 Introduction 2 Proposed Architecture 2.1 Backbone Module 2.2 Simple Spatial Attention 2.3 Connection Module 2.4 Multi-scale Detection and Anchor Assignment 2.5 Multi Boxes Loss Function 3 Dataset and Implementation Setup 4 Experiments and Results 4.1 Ablative Study 4.2 Evaluation on Datasets 4.3 Runtime Efficiency 5 Conclusions References Non-tactile Thumb Tip Measurement System for Encouraging Rehabilitation After Surgery 1 Introduction 2 Related Studies 3 Image Acquisition Using Depth Sensor 4 Thumb Tip Position Measurement by DNN Model 4.1 Extraction of Hand Areas 4.2 Assigning Teacher Signals 4.3 Creating Data Sets 5 Experiments of Thumb Tip Measurement 5.1 Dataset and Prediction Model 5.2 Estimation Results 5.3 Visualization of Thumb Tip Trajectory 5.4 Evaluation 6 User Application for Capturing and Visualizing Measurements 6.1 Depth Image Acquisition 6.2 Outlier Removal for Visualizing Thumb Tip Trajectory 6.3 Creating Mesh Visualizing Thumb Tip Trajectory 6.4 Application Displaying Thumb Tip Trajectory 7 Conclusion References Real-Time Prediction of Future 3D Pose of Person Using RGB-D Camera for Personalized Services 1 Introduction 2 Proposed Method 2.1 Measurement of 3D Pose of Human Using RGB-D Camera 2.2 Training of 3D Motion of Human 2.3 Prediction of 3D Pose of Human 3 Experiments 3.1 Training and Prediction in Experiment 3.2 Experimental Results 4 Conclusion References ROS2-Based Distributed System Implementation for Logging Indoor Human Activities 1 Introduction 1.1 Research Background and Objectives 2 System Overview and System Configuration 2.1 System Overview 2.2 System Configuration 3 System Design 3.1 ROS Package Configuration 3.2 Node Configuration 3.3 Internal Structure of Event Detection and Interpretation Section 4 Experimental Result 4.1 Object Detection Process 4.2 Background Subtraction Extraction Node 4.3 Background Subtraction Analysis Node 4.4 Person Detection Process 4.5 Event Detection and Storing to DB 5 Conclusion References The Virtual Camera Path in 3D Animation 1 Introduction 2 Problem Formulation 2.1 Raw Path 2.2 Segmented and Differentiable Bezier Curves 2.3 Camera Path 3 Path Computation of Virtual Cameras 3.1 The Sampling Strategy 3.2 Segmented and Differentiable Bezier Curve Algorithm 3.3 Control Points and Nonlinear Programming 4 Experimental Results 4.1 Performance 4.2 Comparison 4.3 The Segmentation Results 5 Conclusion References Author Index