Trends and Advancements of Image Processing and Its Applications

Preface
Acknowledgments
Contents
Part I: Recent Trends and Advancements of Image Processing and its Applications
Using Convolutional Neural Networks for Classifying COVID-19 in Computerized Tomography Scans
1 Introduction
1.1 Problem Definition
1.2 Formulation of Hypotheses
1.3 Objective
Specific Objective
1.4 Methodology
2 Literature Review
2.1 Machine Learning
Artificial Neural Networks
Neuroscience Inspiration
Brief History of Neurocomputing
Fundamentals
Learning Paradigms
2.2 Deep Learning
Convolutional Neural Networks
Architecture of Convolutional Neural Networks
2.3 Image Classification
Tomography Scans
Functional Images
PET and Dynamic PET
fMRI
2.4 Tools and Techniques
TensorFlow
Keras
NVIDIA CuDNN
Transfer Learning
Callbacks
3 Experiments
3.1 Environment
3.2 Dataset
3.3 Definition of Models
3.4 Evaluation Metrics
4 Results
5 Conclusion
References
Challenges in Processing Medical Images in Mobile Devices
1 Introduction
2 Background
2.1 Healthcare and Image Processing
2.2 Distributed Computing
3 Related Work
4 The Platform for Distributed Execution
4.1 Selection
4.2 BOINC
5 Application Execution and Transparency for the Programmer
5.1 The Wrapper for Android OS
5.2 Generation of the Binary Code with the ITK Library
5.3 Parallelization of Image Processing
5.4 Performance Results
6 Proper Use of Devices
6.1 SEAS – Simple Energy-Aware Scheduler
6.2 Integrating the Scheduling Strategy into BOINC
7 Conclusions and Future Work
References
Smart Traffic Control for Emergency Vehicles Using the Internet of Things and Image Processing
1 Introduction
1.1 Image Processing
1.2 Internet of Things
2 Literature Review
3 Proposed Work
3.1 Traffic Analyzation at Traffic Signal
3.2 To Update a Load of Traffic to the Main Server
3.3 Updating the Node Weights on the Map
3.4 Update the Path Required to Be Followed by an Emergency Vehicle in Real-Time
3.5 Update the IoT Chip about the Emergency Vehicle
3.6 Set the Timer Accordingly
4 Conclusion and Future Work
References
Combining Image Processing and Artificial Intelligence for Dental Image Analysis: Trends, Challenges, and Applications
1 Introduction
2 An Overview on Digital Dental Imaging
2.1 X-Ray Images
2.2 Intraoral Radiographs
2.3 Extraoral Radiographs
2.4 Computed Tomography and Cone-Beam Computed Tomography
3 Image Processing and Artificial Intelligence for Dental Image Analysis
3.1 Artificial Intelligence Techniques for Image Processing
3.2 Identify Periodontitis
3.3 Detection of Dental Caries
3.4 Image Enhancement
3.5 Other Applications
3.6 Conclusions and Challenge Issues
References
Median Filter Based on the Entropy of the Color Components of RGB Images
1 Introduction
2 Ordering Strategies
2.1 Ordering Techniques
2.2 Vector Ordering Strategies
3 Proposed Ordering Strategy
3.1 Entropy
3.2 Proposed Filter
4 Experiments and Results
4.1 Mean Absolute Error (MAE)
4.2 Implementation of the Proposal
4.3 Experiments and Results
4.4 Results
5 Conclusions and Future Work
References
Deep Learning Models for Predicting COVID-19 Using Chest X-Ray Images
1 Introduction
2 Deep Learning
2.1 Convolutional Neural Network
2.2 Residual Network
2.3 Efficient Networks
3 Related Work
4 Methods and Materials
4.1 Dataset
4.2 Data Preparation
4.3 Data Augmentation
4.4 Environment Setup
5 Results and Discussion
6 Performance Evaluation
7 Conclusion
References
Deep Learning Methods for Chronic Myeloid Leukaemia Diagnosis
1 Scope of Study
1.1 Literature Survey
At the National Level
At the International Level
2 Introduction
2.1 Why Use Deep Learning
3 Progressive Diagnosis of Chronic Myeloid Leukaemia
3.1 Classification Algorithms Used in CML Diagnosis
Support Vector Machines (SVMs)
K-Nearest Neighbour (k-NN) Algorithm
Naïve Bayes
Deep Learning
4 Artificial Neural Networks
5 Convolutional Neural Networks
5.1 Architecture of CNN
6 Typical Application Scenario of Chronic Myeloid Leukaemia
7 Conclusion
References
An Automatic Bean Classification System Based on Visual Features to Assist the Seed Breeding Process
1 Introduction
2 Related Works
3 Proposed Approach
3.1 Bean Image Acquisition
3.2 Seed Segmentation
3.3 Bean Description and Classification
4 Experimental Setup
5 Experimental Results and Discussion
6 Conclusions
References
Supervised Machine Learning Classification of Human Sperm Head Based on Morphological Features
1 Introduction
2 Methodology
2.1 Data Acquisition
2.2 Data Preparation
2.3 Sperm Segmentation
2.4 Spermatozoa Abnormality Classification
2.5 Segmentation and Classification Performance Analysis
3 Results and Discussion
3.1 Segmentation
3.2 Sperm Head Morphology Classification
4 Conclusion and Future Work
References
Future Contribution of Artificial Vision in Methodologies for the Development of Applications That Allow for Identifying Optimal Harvest Times of Medicinal Cannabis Inflorescences in Colombia
1 Introduction
1.1 Inflorescence of Cannabis sp. Maturation and Optimal Harvest Time
Generalities About the Inflorescences of Cannabis sp.
Plant Structures of Interest
Pistils
Trichomes
Harvest, Maturation, and Harvest Time for Cannabis sp. Inflorescences
Types of Maturity
Visual and Touch Indicators of Maturation States
1.2 Artificial Intelligence Applied to Image Recognition
Multilayer Perceptron Artificial Neural Network (MLP ANN)
Convolutional Neural Networks
2 Background on Image Recognition for Similar Applications
3 Methodological Proposal
3.1 Methodology for Capturing Images and Database
3.2 Methodology for the Development of an Application to Recognize the Optimal Harvest Time of the Inflorescences of Cannabis sp.
Definition of Requirements
Software and System Design
4 Exploration in Image Recognition for Inflorescences of Cannabis sp. Using Artificial Neural Networks
4.1 Construction of the Artificial Neural Network Training Set
4.2 Elaboration and Training of Artificial Neural Networks for Estimating the Maturity State of Cannabis sp. Inflorescences
5 Conclusions and Future Agenda
References
Detection of Brain Tumor Region in MRI Image Through K-Means Clustering Algorithms
1 Introduction
2 Literature Review
3 Proposed Method
4 K-Means Clustering Algorithm
4.1 Mathematical Representation
4.2 Algorithm
4.3 Pre-processing
4.4 Post-processing Comprises
5 Simulation and Results
6 Conclusion
References
Estimation of Human Posture Using Convolutional Neural Network Using Web Architecture
1 Introduction
1.1 Pose Estimation
1.2 Pose Estimation Metrics
2 Literature Review
2.1 Background and Techniques
3 Modules of the Proposed System
3.1 Human Pose Extraction
Convolutional Neural Network
Monitoring of Optical Flow
Algorithm of Time Series Data Alignment
Description of Framework
4 Conclusion and Future Scope
4.1 Conclusion
4.2 Scope of the Future
References
Histogram Distance Metric Learning to Diagnose Breast Cancer using Semantic Analysis and Natural Language Interpretation Methods
1 Introduction
2 NLP Levels
2.1 Phonology
2.2 Morphology
2.3 Lexical
2.4 Syntactic
2.5 Semantic
3 Natural Language Generation (NLG)
3.1 Speaker and Generator
3.2 Representation Elements and Rates
3.3 Theological Organization
3.4 Program or Speaker
4 Medicine
5 Breast Cancer
5.1 Categories of Breast Cancer
5.2 Development of Cancer
6 Mammography Screening
6.1 Mammography Diagnosis
6.2 Mammography Online
6.3 Screen Film Mammography
7 Conclusion
References
Human Skin Color Detection Technique Using Different Color Models
1 Introduction
2 Color Models
2.1 Red, Green, and Blue (RGB) Model
2.2 Hue, Saturation, and Value (HSV) Model
2.3 YCbCr Model
2.4 CMYK (Cyan, Magenta, Yellow, and Black) Model
3 Image Processing and Color Models
3.1 Pseudo-color Image Processing
3.2 Gray Level to Color Transformation Method
3.3 Full-Color Image Processing
3.4 Color Transformation
4 Related Work
5 Proposed Methodology
5.1 Proposed Skin Detection Algorithm
5.2 Restriction of Different Color Spaces
6 Results and Discussion
6.1 Comparative Analysis of Different Color Models
6.2 ROC Analysis
6.3 Experimental Results of Sample Image
7 Conclusion
References
A Study of Improved Methods on Image Inpainting
1 Introduction
2 Comparative Study
2.1 Image Inpainting
2.2 Sparse Representation
2.3 Markov Random Field Modeling and Multiscale Graph Cuts
2.4 Neural Networks and GAN
2.5 Texture Synthesis
2.6 Feature Distribution
3 Conclusion
References
Index