Recent Trends in Image Processing and Pattern Recognition: 4th International Conference, RTIP2R 2021, Msida, Malta, December 8-10, 2021, Revised ... in Computer and Information Science, 1576)

Preface
Organization
Contents
Healthcare: Medical Imaging and Informatics
Cleaning Highly Unbalanced Multisource Image Dataset for Quality Control in Cervical Precancer Screening
	1 Introduction
	2 Image Data
	3 Method
	4 Experiments and Discussions
	5 Conclusion and Future Work
	References
Detection of Male Fertility Using AI-Driven Tools
	1 Introduction
	2 Related Works
	3 Methods
		3.1 Features Detailing
		3.2 Feature Selection
		3.3 Feature Selection
		3.4 The Workflow of Proposed Model
	4 Experiments
		4.1 Data
		4.2 Evaluation Protocols
		4.3 Results
	5 Conclusion
	References
An Empirical Study of Vision Transformers for Cervical Precancer Detection
	1 Introduction
	2 Related Work
	3 Experiments
		3.1 Data Preparation
		3.2 Architecture
		3.3 Training Setup
	4 Results and Discussions
	5 Conclusion and Future Work
	References
CheXNet for the Evidence of Covid-19 Using 2.3K Positive Chest X-rays
	1 Introduction
	2 Chest X-ray Dataset
	3 CheXNet and Implementation
	4 Experimental Results
		4.1 Evaluation Protocol and Performance Metrics
		4.2 Results Analysis
		4.3 Comparative Study
	5 Conclusion and Future Work
	References
An Enhanced Deep Convolution Neural Network Model to Diagnose Alzheimer's Disease Using Brain Magnetic Resonance Imaging
	1 Introduction
	2 Proposed Methodology
		2.1 Data Pre-processing
		2.2 Deep CNN: Architecture and Implementation
		2.3 Hyper Parameters in Deep CNN
	3 Experiments
		3.1 Dataset
		3.2 Validation Protocol and Evaluation Metrics
		3.3 Results and Discussion
	4 Conclusion
	References
Automatic Knee Osteoarthritis Stages Identification
	1 Introduction
	2 Literature Review
	3 Methodology
		3.1 Data Acquisition
		3.2 Left and Right Knee Segmentation
		3.3 ROI Detection
		3.4 DNN-Based Classification System
	4 Experimental Setup and Result Discussion
		4.1 Comparison of OACnet with Other DNN-Based Models
	5 Conclusion
	References
Stacked Dark COVID-Net: A Multi-class Multi-label Classification Approach for Diagnosing COVID-19 Using Chest X-Ray Images
	1 Introduction
	2 Literature Review
	3 Proposed Methodology
		3.1 Image Processing
		3.2 Layered DarkNet (LDN)
		3.3 Novel Stacked Dark COVID-Net
	4 Experimental Setup and Discussions
		4.1 Experimental Setup
		4.2 Dataset Description
	5 Conclusions
	References
Image Augmentation for Improving Automated Eligibility-Classification for Cervical Precancer Ablation Treatment
	1 Introduction
	2 Methods
		2.1 Dataset and Annotation
		2.2 CycleGan
		2.3 Customized CNN
	3 Experiments
		3.1 Augmented Images
		3.2 Treatability Classification Performance
		3.3 Comparing the Visualization
	4 Conclusion
	References
Osteoarthritis Detection Using Densely Connected Neural Network
	1 Introduction
	2 Related Work
	3 Architecture of Dense Net
	4 Design of OA Detection and Classification Model
		4.1 Input Image Acquisition
		4.2 Pre-processing
		4.3 Feature Extraction
		4.4 Image Classification
	5 Results and Discussion
	6 Conclusion
	References
Generic Foreign Object Detection in Chest X-rays
	1 Introduction
	2 Related Works
	3 YOLOv4, Architecture and Implementation
	4 Experiments
		4.1 Dataset
		4.2 Results and Analysis
	5 Conclusions
	References
Mammogram Mass Classification: A CNN-Based Technique Applied to Different Age Groups
	1 Contribution Outline
		1.1 Traditional Classifiers
		1.2 CNN Architecture
	2 Experiments
		2.1 Dataset and Pre-processing
		2.2 Age Grouping
		2.3 Evaluation Metrics
		2.4 Evaluation Strategy and System Configuration
		2.5 Results and Analysis
		2.6 Comparative Analysis with Previous Relevant Work
	3 Conclusions and Future Work
	References
Computer Vision and Pattern Recognition
Complex Object Detection Using Light-Field Plenoptic Camera
	1 Introduction
	2 Plenoptic Camera Calibration
	3 Methods
		3.1 Feature Detection
		3.2 Feature Matching
		3.3 Transform Model Estimation
		3.4 Results
		3.5 Conclusions
	References
Real-Time Face Recognition for Organisational Attendance Systems
	1 Introduction
	2 Literature Survey
	3 Methodology
	4 Experimental Results and Discussion
	5 Conclusion
	References
Harnessing Sustainable Development in Image Recognition Through No-Code AI Applications: A Comparative Analysis
	1 Introduction
	2 Literature Review
		2.1 Previous Research
	3 Methodology
		3.1 Dataset
		3.2 Hardware
		3.3 Software
		3.4 ResNet – 50 V2
		3.5 MobileNetV2
	4 Results
	5 Discussion
	References
Evaluating Performance of Adam Optimization by Proposing Energy Index
	1 Introduction
	2 Related Work
		2.1 AdaGrad
		2.2 AdaDelta
		2.3 RMSProp
		2.4 Adam
		2.5 Energy Index Based Optimization Method (EIOM)
	3 Introducing EI to Adam
	4 Experiments and Analysis
		4.1 Logistic Regression
		4.2 Support Vector Machine
	5 Conclusion and Future Direction
	References
An Alignment-Free Fingerprint Template Protection Technique Based on Minutiae Triplets
	1 Introduction
	2 Literature Survey
	3 Proposed Technique
		3.1 Non-invertible Feature Extraction
		3.2 Secured Template Generation
		3.3 Security Enhancement of the Generated Template
		3.4 Matching
	4 Experimental Analysis
		4.1 Revocability
		4.2 Diversity
		4.3 Security
		4.4 Performance
	5 Conclusion
	References
Early Prediction of Complex Business Processes Using Association Rule Based Mining
	1 Introduction
	2 Related Work
	3 Association Rule Mining
	4 The CBA-Based Prediction Model
	5 Experimental Results
		5.1 Comparative Analysis
	6 Conclusion
	References
A Framework for Masked-Image Recognition System in COVID-19 Era
	1 Introduction
		1.1 Objectives
	2 Literature Review
	3 Problem Formulation
	4 Proposed Solutions
		4.1 Programming Languages and Tools
		4.2 Web Application Frameworks
		4.3 Computer Vision Frameworks
		4.4 Algorithms
		4.5 Approaches
	5 Specification and Design
		5.1 System Requirements
		5.2 Principle System Components and Requirements
	6 Implementations and Result Analysis
		6.1 Home Component
		6.2 Camera Component
		6.3 Login Component
		6.4 Profile Component
		6.5 Results Analysis
	7 Conclusions and Future Work
	References
A Deep-Learning Based Automated COVID-19 Physical Distance Measurement System Using Surveillance Video
	1 Introduction
	2 Related Works
	3 Proposed Method
		3.1 Human Detection
		3.2 Human Tracking
		3.3 Distance Measurement
	4 Experimental Results and Discussions
		4.1 Dataset and Training Details
		4.2 Results and Discussions
	5 Conclusion
	References
Face Mask Detection Using Deep Hybrid Network Architectures
	1 Introduction and Problem Understanding
	2 Related Works
	3 Dataset and Data Preprocessing
		3.1 Data Description
		3.2 Resizing
	4 Description of the Network Architectures and Machine Learning Algorithms
		4.1 ResNet50
		4.2 VGG16
		4.3 Support Vector Machines
		4.4 GBDecision Trees
		4.5 Neural Networks
		4.6 Hybrid Models
	5 Proposed Model: ResNet50 + SVM
	6 Results and Discussion
	7 Conclusion
	References
A Super Feature Transform for Small-Size Image Forgery Detection
	1 Introduction
	2 Related Works and Problem Context
		2.1 Preliminaries of Super Resolution Algorithms
		2.2 Review of Forgery Detection Algorithms
	3 Proposed Detection Model
		3.1 Super Feature Transform (SFT)
		3.2 Feature Filtering and Forged Region Localization
	4 Experimentation and Analysis
		4.1 Evaluation of Super Feature Transform
		4.2 Detection Results with Post-processing Actions
	5 Conclusion
	References
Document Analysis and Recognition
UHTelHwCC: A Dataset for Telugu Off-line Handwritten Character Recognition
	1 Introduction
	2 Related Work
	3 UHTelHwCC Creation
	4 UHTelHwCC Characteristics
		4.1 Sample Distribution
		4.2 Data Splits
		4.3 Writers Details
	5 Experiments
	6 Conclusion
	References
Inflectional and Derivational Hybrid Stemmer for Sentiment Analysis: A Case Study with Marathi Tweets
	1 Introduction
	2 Related Work
	3 Marathi Morphology
	4 Marathi Stemming Algorithm
		4.1 Data Collection
		4.2 Cleaning
		4.3 Stop Word Removal
		4.4 Marathi Subjective Lexicon (Adjective-Adverb)
		4.5 Stemming
	5 Experiments and Results
		5.1 Performance of the Stemmer
		5.2 Tweet Classification and Stemmer Validation
		5.3 Comparison with Existing Marathi Stemmers
	6 Conclusion
	References
Adaptive Threshold-Based Database Preparation Method for Handwritten Image Classification
	1 Introduction
	2 Related Works
	3 Methodology
		3.1 Deep Neural Network
		3.2 Feature Extraction with Adaptive Threshold
		3.3 Classification and Recognition
	4 Experimental Setup and Evaluation Protocol
		4.1 Dataset and Pre-processing
		4.2 Evolution Protocol
	5 Conclusion and Future Scope
	References
A Graph-Based Holistic Recognition of Handwritten Devanagari Words: An Approach Based on Spectral Graph Embedding
	1 Introduction
	2 Proposed Model
		2.1 Pre-processing
		2.2 Proposed Graph Extraction/Representation
		2.3 Edge Connecting, Labelling and Selection of Graph Associated Matrices
		2.4 Spectral Decomposition of Word Graph Associated Matrices
	3 Adequacy of the Features
	4 Experimentation
		4.1 Dataset
		4.2 Experimental Setup
		4.3 Experimental Results
	5 Comparative Analysis
	6 Conclusion and Future Work
	References
Signal Processing and Machine Learning
Imagined Object Recognition Using EEG-Based Neurological Brain Signals
	1 Introduction
	2 Proposed Methodology
		2.1 Signal Preprocessing
		2.2 Feature Extraction
		2.3 Random Forest (RF)
	3 Experiments and Results
		3.1 Dataset Description and Experiment Protocol
		3.2 Experimental Results
	4 Conclusion and Future Scope
	References
A Fast and Efficient K-Nearest Neighbor Classifier Using a Convex Envelope
	1 Introduction
	2 Related Work
	3 Method
	4 Experiments
		4.1 Data Collections
		4.2 Experimental System
		4.3 Results
	5 Conclusion
	References
Single Channel Speech Enhancement Using Masking Based on Sinusoidal Modeling
	1 Introduction
	2 Signal Model and Notation
	3 Sinusoidal Model Based Speech Enhancement
		3.1 Finding of Peaks of the Power Spectrum
		3.2 Peak Matching Algorithm
		3.3 Wiener Filter Gain Function
		3.4 Synthesis System
	4 Results and Discussion
	5 Conclusion
	References
Extraction of Temporal Features on Fibonacci Space for Audio Based Vehicle Classification
	1 Introduction
	2 Feature Extraction
		2.1 Fibonacci Series
	3 Preprocessing: Segmentation and Windowing
		3.1 Algorithm
		3.2 Skewness
		3.3 Kurtosis
		3.4 Entropy
	4 Results and Discussion
		4.1 Database Used
		4.2 Convolution Neural Network (CNN)
		4.3 Performance Analysis
	5 Conclusion
	References
An Improved Technique for Preliminary Diagnosis of COVID-19 via Cough Audio Analysis
	1 Introduction
	2 Background
		2.1 Laboratory-Based Testing
		2.2 Radiography Testing
		2.3 Cough-Based Testing
	3 Methodology
		3.1 Proposed Architecture
		3.2 COVID-19 Cough Dataset
		3.3 Data Augmentation
		3.4 Feature Extraction
	4 Model Architecture
		4.1 Multilayer Perceptron
		4.2 Convolutional Neural Networks
		4.3 Recurrent Neural Networks with Long Short-Term Memory
		4.4 Support Vector Machines
	5 Results
		5.1 Accuracy
		5.2 Precision
		5.3 Recall
	6 Conclusion
	A  Appendix
		A.1  Softmax
		A.2  ReLU
	References
Satellite Imaging and Remote Sensing
Agricultural Field Analysis Using Satellite Hyperspectral Data and Autoencoder
	1 Introduction
	2 Related Work
	3 Proposed Methodology
		3.1 Dataset Preparation
		3.2 Feature Extraction
		3.3 Clustering
		3.4 Classification
	4 Results
	5 Conclusion
	6 Future Work
	References
Development of NDVI Prediction Model Using Artificial Neural Networks
	1 Introduction
	2 Literature Survey
	3 Study Area
	4 Dataset
	5 Methodology
		5.1 Preprocessing
		5.2 Normalised Difference Vegetation Index (NDVI)
		5.3 Raster to Point Module
		5.4 Grid of Tiles Module
		5.5 Point Clipping Module
		5.6 Data Cleaning Module
		5.7 Architecture of the ANN
		5.8 Vegetation Cover Prediction Module Using MATLAB
		5.9 Data Post-processing Module
		5.10 Predicted Points Module
		5.11 Point of Raster
		5.12 Predicted Map
	6 Result and Discussions
	7 Conclusions
	References
Time Series Forecasting of Soil Moisture Using Satellite Images
	1 Introduction
	2 Literature Review
	3 Methodology
		3.1 Hardware Setup
		3.2 The Proposed Pipeline
		3.3 Dataset Overview
		3.4 Data Preprocessing
		3.5 Dimensionality Reduction
		3.6 Time Series Forecasting
	4 Results
		4.1 Evaluation Metrics
		4.2 Comparative Analysis of AutoEncoders and PCA for Dimensionality Reduction
		4.3 Comparative Analysis of Different LSTM Architectures for Forecasting
	5 Conclusions and Future Scope
	References
Author Index