Applied Computer Vision and Soft Computing with Interpretable AI

Cover
Half Title
Title Page
Copyright Page
Table of Contents
Preface
Editors
Contributors
Chapter 1: Improved Healthcare Systems Using Artificial Intelligence: Technology and Challenges
	1.1 Introduction
	1.2 Motivation
	1.3 Literature Review
	1.4 Technology in Healthcare
		1.4.1 Accurate Cancer Diagnosis
		1.4.2 Premature Detection of Lethal Blood Diseases
		1.4.3 Customer Service Chatbots
		1.4.4 Treatment of Odd Diseases
		1.4.5 Automation of Repetitive Jobs
		1.4.6 Handling and Supervision of Medical Records
		1.4.7 Development of New Medicines
		1.4.8 Robot-assisted Surgery
		1.4.9 Automation of Medical Image Diagnoses
	1.5 Challenges and Solutions
		1.5.1 AI Bias
		1.5.2 Personal Security
		1.5.3 Transparency
		1.5.4 Data Formats
		1.5.5 Societal Acceptance/Human Factors
	1.6 Conclusion and Future Scope
	References
Chapter 2: A Brain MRI Segmentation Method Using Feature Weighting and a Combination of Efficient Visual Features
	2.1 Introduction: Background and Driving Forces
	2.2 Proposed Framework
		2.2.1 Approach Overview
		2.2.2 Preprocessing
		2.2.3 Feature Extraction
		2.2.4 Clustering Step
		2.2.5 Post-processing
	2.3 Experiments
		2.3.1 Dataset
		2.3.2 Performance Metrics
			Experiment 1: The Analysis of Extracted Features
			Experiment 2: The Impact of the Feature Weighting Strategy
			Experiment 3: The Proposed Method vs. Other Methods
	2.4 Conclusion
	Note
	References
Chapter 3: Vision Based Skin Cancer Detection: Various Approaches with a Comparative Study
	3.1 Introduction: Background and Driving Forces
		3.1.1 Problem Formulation and Motivation
			3.1.1.1 Proposed Solution
			3.1.1.2 Scope of the Proposed Solution
		3.1.2 Review of the Literature
			3.1.2.1 Image Preprocessing and Enhancement
			3.1.2.2 Image Segmentation
			3.1.2.3 Feature Extraction
			3.1.2.4 Classification
		3.1.3 Algorithmic View with Implementation Details
			3.1.3.1 Preprocessing
			3.1.3.2 Segmentation
			3.1.3.3 Feature Extraction
			3.1.3.4 Classification
		3.1.4 Results and Discussion
			3.1.4.1 Performance of Eight-bins
			3.1.4.2 Performance of CNN
	3.2 Conclusion, Take-aways, and Future Directions
	References
Chapter 4: MentoCare: An Improved Mental Healthcare System for the Public
	4.1 Introduction
	4.2 Related Work
	4.3 Proposed Methodology
	4.4 Results and Discussion
	4.5 Conclusion and Future Scope
	References
Chapter 5: An Employee Health Monitoring System Using Wireless Body Area Networks and Machine Learning
	5.1 Introduction
	5.2 Literature Survey
	5.3 MI Theory
		5.3.1 STEMI
		5.3.2 NSTEMI
		5.3.3 Angina
	5.4 Proposed Methodology
		5.4.1 Data creation
		5.4.2 Authentication
		5.4.3 Disease Prediction System (DPS)
	5.5 Algorithm for DLNN
	5.6 UML and Working
	5.7 Working on the Proposed Project
		5.7.1 Admin Interface
		5.7.2 Medical Professional Interface
		5.7.3 Employer Interface
	5.8 Conclusion
	Appendix A
	References
Chapter 6: Monitoring Operational Parameters in the Manufacturing Industry Using Web Analytical Dashboards
	6.1 Introduction
	6.2 Challenges
	6.3 Literature Review
	6.4 Methodology
	6.5 Datasets
	6.6 Experimental Investigation
		6.6.1 Need of Data Analytics
			6.6.1.1 Steps of Data Analytics
			6.6.1.2 Types of Data Analytics
			6.6.1.3 Benefits of Data Analytics [ 23–29 ]
	6.7 Tools and Technologies Used
	6.8 Results and Discussion
	6.9 Results
		6.9.1 Daily Production Report Dashboard
		6.9.2 MIS Production Report Dashboard (Sheet 1: FC-Machine Shop)
		6.9.3 MIS Production Report Dashboard (Sheet 2: FC-Operations in House Production)
	6.10 Future Directions
	Acknowledgment
	References
Chapter 7: Concurrent Line Perpendicular Distance Functions for Contour Point Analysis
	7.1 Introduction
	7.2 Background
	7.3 Shape Descriptor
	7.4 Scale Invariant Features
	7.5 Experiments and Analysis
		7.5.1 Kimia’s Dataset
		7.5.2 MPEG-7 Dataset
	7.6 Conclusions
	References
Chapter 8: A Resemblance of Convolutional Neural Network Architectures for Classifying Ferrograph Images
	8.1 Introduction
	8.2 Dataset
	8.3 Transfer Learning and Fine Tuning
	8.4 Hardware and Convolutional Neural Network Architectures
		8.4.1 VGG
		8.4.2 ResNet
		8.4.3 InceptionV3
		8.4.4 Xception
		8.4.5 MobileNet
		8.4.6 DenseNet
		8.4.7 MobileNetV2
		8.4.8 EfficientNet
		8.4.9 ConvNeXt
	8.5 Model Configuration and Training
	8.6 Results
	8.7 Conclusion
	References
Chapter 9: The Role of Artificial Intelligence and the Internet of Things in Smart Agriculture towards Green Engineering
	9.1 Introduction
	9.2 Artificial Intelligence in Agriculture
	9.3 Precision Agriculture Artificial Intelligence
		9.3.1 Geographic Information System (GIS)
		9.3.2 Autosteer
	9.4 Agricultural Robotics and Drones
		9.4.1 Harvest CROO Robotics
		9.4.2 Robot Drone Tractors
		9.4.3 Farm Bots
		9.4.4 Autonomous Tractors
		9.4.5 Unmanned Aerial Vehicles (UAVs)
	9.5 Image-based Insight Generation
	9.6 Artificial Intelligence in Management Accounting
	9.7 Agriculture and the Internet of Things
	9.8 The Precision Farming Internet of Things (IoT)
		9.8.1 Agriculture Sensors
		9.8.2 Communication in Agriculture
	9.9 The Internet of Things Cloud
		9.9.1 Climate Change
		9.9.2 Smart Greenhouses
		9.9.3 Internet of Things-based Tractors
	9.10 Challenges with the Internet of Things
		9.10.1 Future Scope of the Internet of Things in Agriculture
	9.11 Integrating Artificial Intelligence and the Internet of Things in Agriculture
	9.12 Applications of Artificial Intelligence and the Internet of Things in Agriculture
	9.13 Conclusion
	References
Chapter 10: Intuitionistic Fuzzy Hypergraphs and Their Operations
	10.1 Introduction
	10.2 The Literature Review
	10.3 Preliminaries
	10.4 Different Types of Operations with Respect to IFHGs
		10.4.1 Complement of an IFHG
		10.4.2 Union of Two IFHGs
		10.4.3 Intersection of Two IFHGs
		10.4.4 Ring Sum of Two IFHGs
		10.4.5 Join of Two IFHGs
		10.4.6 Cartesian Product of Two IFHGs
		10.4.7 Composition of Two IFHGs
	10.5 Summary
	List of Abbreviations
	References
Chapter 11: Spammer Detection Based on a Heterogeneous Multiple-mini-graph Neural Network
	11.1 Introduction
	11.2 Literature Review
		11.2.1 Existing Work
		11.2.2 Summary of the Literature
	11.3 Graph Terminologies
		11.3.1 Graph Neural Networks
		11.3.2 Graph Convolutional Networks
		11.3.3 Heterogeneous GNNs
		11.3.4 Vanilla Feature Embedding
		11.3.5 Random Walk
	11.4 Proposed Spammer Detection Methodology
		11.4.1 Hypergraph Generation
		11.4.2 Heterogeneous Graph Convolution
		11.4.3 Model Training and Analysis
			11.4.3.1 Model Training
			11.4.3.2 Model Analysis
	11.5 Experimental Setup and Results
		11.5.1 Parameters Defined
		11.5.2 Experimental Setting
			11.5.2.1 Preprocessing Input
		11.5.3 Performance Analysis
		11.5.4 Performance Comparison
	11.6 Conclusion
	References
Chapter 12: Spam Email Classification Using Meta-heuristic Algorithms
	12.1 Introduction
	12.2 Related Work
	12.3 Proposed System Architecture
		12.3.1 Pre-processing
		12.3.2 Horse Herd Optimization Algorithm
		12.3.3 Multi-objective Opposition-based Binary HOA
		12.3.4 Spam Detection Using MOBHOA
	12.4 Results Analysis
	12.5 Conclusion
	Conflict of Interest
	References
Chapter 13: A Blockchain Model for Land Registration Properties in Metro Cities
	13.1 Introduction
		13.1.1 Land Registration Types
		13.1.2 Issues or Challenges in Land Registry, Maharashtra, India
		13.1.3 Use of Blockchain Technology for These Issues
		13.1.4 Structure of Blockchains
		13.1.5 The Various Kinds of Agreement Conventions Utilized for Approving Exchanges on the Blockchain
	13.2 Current Land Registration Procedure
		13.2.2 Measures which Should be Taken to Avoid Bad Land Deals
		13.2.3 Types of Blockchain Technology for Land Registration
		13.2.4 Hybrid Blockchains
		13.2.5 Case Study: Gujarat Land Registration
	13.3 Proposed Hybrid Blockchain Model for Land Registry in Maharashtra, Pune
	13.4 Future Scope and Conclusion
	References
Chapter 14: A Review of Sentiment Analysis Applications and Challenges
	14.1 Introduction
	14.2 Sentiment Analysis: An Overview
		14.2.1 Level of Aspect
		14.2.2 Level of Sentence
		14.2.3 Level of Document
	14.3 Challenges
		14.3.1 Unstructured Data
		14.3.2 Aspect Identification
		14.3.3 Sentiment Identification
			14.3.3.1 Sentiment Recognition Using Supervised Methods
			14.3.3.2 Sentiment Recognition Using Unsupervised Methods
			14.3.3.3 Lexical Analysis for Sentiment Recognition
		14.3.4 Topic Model-Based Approaches
	14.4 Applications of Sentiment Analysis
		14.4.1 Business Intelligence
		14.4.2 Review Analysis
		14.4.3 The Stock Market
		14.4.4 Healthcare
		14.4.5 Behavior Analysis
		14.4.6 Social Media Analysis
		14.4.7 Email Mining
	14.5 Performance Evaluation Parameters
	14.6 Conclusions
	14.7 Further Research
	Conflict of Interest
	References
Chapter 15: Handling Skewed Datasets in Computing Environments: The Classifier Ensemble Approach
	15.1 Building a Classifier Ensemble
		15.1.1 Diversity among Different Classifiers
	15.2 Base Classifiers for Classifier Ensembles
		15.2.1 Support Vector Machine (SVM)
		15.2.2 Decision Tree
		15.2.3 Multilayer Perceptron (MLP)
	15.3 Ensemble Combination Strategy
		15.3.1 Classifier Fusion
			15.3.1.1 Voting
		15.3.2 Classifier Selection
	15.4 Concluding Remarks
	References
Chapter 16: Diagnosis of Dementia Using MRI: A Machine Learning Approach
	16.1 Introduction
		16.1.1 Alzheimer’s Disease (AD)
			16.1.1.1 Early-stage Alzheimer’s (Mild)
			16.1.1.2 Middle-stage Alzheimer’s (Moderate)
			16.1.1.3 Late-stage Alzheimer’s (Severe)
		16.1.2 Vascular Dementia (VD)
		16.1.3 Lewy Body Dementia (LBD)
		16.1.4 Frontotemporal Dementia (FTD)
		16.1.5 Mixed Dementia
	16.2 Literature Survey
	16.3 Algorithmic Survey
		16.3.1 Support Vector Machine (SVM)
		16.3.2 Convolutional Neural Network (CNN)
		16.3.3 Naïve Bayes
		16.3.4 Decision Tree
		16.3.5 Logistic Regression
		16.3.6 Multilayer Perceptron (MLP)
		16.3.7 Voting Based Classifiers
		16.3.8 K-Nearest Neighbour
		16.3.9 Extreme Gradient Boosting (XGB)
		16.3.10 Kernel Support Vector Machine
		16.3.11 Radial Basis Function
		16.3.12 Gaussian Mixture Model
	16.4 Proposed Methodology
		16.4.1 Introduction
		16.4.2 Dataset
		16.4.3 Data Pre-processing
		16.4.4 Visualizing Data
		16.4.5 Feature Extraction
		16.4.6 Applying ML and DL Techniques
		16.4.7 Classification
		16.4.8 Prediction
	16.5 Results
	16.6 Conclusion and Future Work
	Acknowledgement
	References
Chapter 17: Optimized Student’s Multi-Face Recognition and Identification Using Deep Learning
	17.1 Introduction
	17.2 The Literature
		17.2.1 Technical Survey
		17.2.2 Non-Technical Survey
	17.3 Common Findings from the Survey
	17.4 Results and Discussion
	17.5 Conclusion
	References
Chapter 18: Impact of Fake News on Society with Detection and Classification Techniques
	18.1 Introduction
	18.2 Research Methodology and Algorithm Design
		18.2.1 Machine Learning Models
		18.2.2 Machine Learning Model Evaluation
		18.2.3 Algorithm Design for Proposed Model
	18.3 Results and Discussion
	18.4 Conclusion
	References
Chapter 19: Neurological Disorder Detection Using Computer Vision and Machine Learning Techniques
	19.1 Introduction
	19.2 Literature Review
	19.3 Methodology
		19.3.1 Thresholding
		19.3.2 Segmentation
		19.3.3 Edge Based Segmentation Method
		19.3.4 Region-Dependent Segmentation Approach
		19.3.5 Convolution Neural Networks (CNNs)
		19.3.6 KNN Algorithm
	19.4 System Architecture
	19.5 Results and Discussion
	19.6 Conclusion
	References
Chapter 20: Deep Learning for Tea Leaf Disease Classification: Challenges, Study Gaps, and Emerging Technologies
	20.1 Introduction
	20.2 Motivation
	20.3 Literature Review
	20.4 Challenges in DL for Tea Leaf Disease Classification
		20.4.1 Variations in Symptoms
		20.4.2 Interclass Similarities
		20.4.3 Image Background
		20.4.4 Other Problems
	20.5 A Review of Recent CNN Architectures for Tea Leaf Disease Classification
		20.5.1 GoogleNet
		20.5.2 AlexNet
		20.5.3 VGG16
		20.5.4 ResNet50
		20.5.5 LeafNet
		20.5.6 MergeModel
		20.5.7 Xiaoxiao SUN1’s CNN Architecture
		20.5.8 LeNet
	20.6 Trending Models and Techniques Used in This Field
	20.7 Conclusion
	References
Index