Recent Trends in Image Processing and Pattern Recognition: Third International Conference, RTIP2R 2020, Aurangabad, India, January 3–4, 2020, Revised ... in Computer and Information Science)

Preface
Organization
Contents – Part I
Contents -- Part II
Computer Vision and Applications
Detection of Road Sign Using Edge Detection Method
	1 Introduction
	2 Proposed Methodology
		2.1 Color Extraction
		2.2 Shape Extraction
		2.3 Sign Detection
	3 Implementation
	4 Observations and Results
	5 Conclusion
	References
Fuzzy Approach to Evaluate Performance of Teaching Staff in Technical Institutions
	1 Introduction
		1.1 Phases of the Fuzzy Logic
		1.2 Different Fuzzy Membership Functions
	2 Proposed Method
	3 Experimental Result
	4 Conclusion
	References
Detection and Learning of Color Objects Using Kernelized Support Correlation Filter
	1 Introduction
	2 Support Correlation Filtering
		2.1 Class Labels
		2.2 Similarities or Dissimilarities with Existing CF-Based Trackers
		2.3 Kernelized SCF
		2.4 Optimization Algorithm
	3 Experimental Result
		3.1 Experiments on Image Sequences
	4 Conclusion
	References
Recent Advances in IoT Based Smart Object Detection and Its Authentication by BlockChain Approaches
	1 Introduction
	2 Literature Review on IoT Based Object Detection
	3 Literature Review on BlockChain Based Security and Privacy
	4 Literature Review on BlockChain Based Video Processing
	5 Conclusion
	References
MINU-EXTRACTNET: Automatic Latent Fingerprint Feature Extraction System Using Deep Convolutional Neural Network
	1 Introduction
	2 Related Work on Fingerprint Matching
		2.1 Pre-ProcessNet
		2.2 ExtractNet
	3 Results
		3.1 Datasets
	4 Conclusion and Future Scope
	References
Texture Based Material Classification Using Gabor Filter
	1 Introduction
	2 Literature Review
	3 Features
		3.1 Color Feature
		3.2 Luminance Feature
		3.3 Texture Feature
	4 Features Extraction
		4.1 Color Feature Extraction
		4.2 Luminance Feature Extraction
		4.3 Feature Extraction Using GLCM
		4.4 Feature Extraction Using Gabor
	5 K-Nearest Neighbors Classifier
	6 Result and Discussions
	7 Conclusions
	References
Optimization of Face Retrieval and Real Time Face Recognition Systems Using Heuristic Indexing
	1 Introduction
		1.1 Size of Search Space
		1.2 Complexity of Recognition Algorithm
		1.3 Face Recognition Systems
	2 Literature Survey
	3 Proposed Approach
		3.1 Preprocessing
		3.2 Localization of Key–Pixels
		3.3 Skin Color and Skin Complexion
		3.4 Tolerance Range
	4 Database Used
	5 Face Recognition
	6 Evolution Metrics
		6.1 Comparative Results
	7 Platform Used
	8 Conclusion
	References
Estimation of Human Age and Gender Based on LBP Features Using Two Level Decision by SVM
	1 Introduction
	2 Related Work
	3 Proposed Methodology
		3.1 Image Acquisition
		3.2 Preprocessing
		3.3 Segmentation
	4 Feature Extraction
	5 Age Estimation
	6 Experimental Results
	7 Conclusion
	References
Assistive Technologies for Visually Impaired Persons Using Image Processing Techniques – A Survey
	1 Introduction
	2 Existing ETA Products
		2.1 Ultrasonic Sensor or Sonar Based systems
		2.2 Infrared Sensors based Systems
	3 Literature Review
	4 General Discussion About Assistive Technologies
	5 Shortcomings Observed in Current Assistive Devices
	6 Proposed System: Use of Image Processing in Assistive Technology Devices
	7 Listing of Some Recent Research and Development Projects in Assistive Technology
	8 Conclusion
	References
Script Identification of Movie Titles from Posters
	1 Introduction
	2 Literature Study
	3 Dataset Collection
	4 Proposed Method
		4.1 Text Localization
		4.2 Movie Title Extraction
		4.3 Feature Consideration
		4.4 Classification
	5 Experiment
		5.1 Result and Discussion
		5.2 Error Analysis
	6 Conclusion
	References
Ensemble of Nested Dichotomies for Author Identification System Using Similarity-Based Textual Features
	1 Introduction
	2 Proposed Methodology
		2.1 Data
		2.2 Pre-processing
		2.3 Feature
		2.4 Ensemble of Nested Dichotomies
	3 Experiments
		3.1 Our Results
		3.2 Comparison with Other Classifiers
		3.3 Comparison with Existing Works
	4 Conclusion
	References
Feature Combination of Pauli and H/A/Alpha Decomposition for Improved Oil Spill Detection Using SAR
	1 Introduction
	2 Dataset and Tools
	3 Proposed Methodology
	4 Pauli Decomposition
	5 H/A/Alpha Decomposition
	6 Proposed Feature Combination Technique
	7 Result Analysis
	8 Conclusion
	References
A Fast and Efficient Convolutional Neural Network for Fruit Recognition and Classification
	1 Introduction
	2 Existing Work
		2.1 Image Classification
		2.2 Deep Learning
		2.3 Fruit Classification
	3 Proposed Method
	4 Experimental Results
		4.1 Dataset
	5 Performance Comparison
		5.1 Training Performance
		5.2 Testing Performance
	6 Conclusion
	References
Copy-Move Image Forgery Detection Using Discrete Wavelet Transform
	1 Introduction
	2 Related Work
	3 Proposed Methodology
		3.1 Preprocessing
		3.2 Feature Extraction
		3.3 Matching Process
	4 Evaluation
	5 Result and Discussion
	6 Conclusion
	References
A Comprehensive Survey of Different Phases for Involuntary System for Face Emotion Recognition
	1 Introduction
	2 Face Acquisition and Registration
	3 Pre-processing and Dimension Reduction Techniques
	4 Datasets
	5 Feature Extraction and Learning
		5.1 Appearance Based Learning
		5.2 Shape Based Learning
		5.3 Neural Networks
	6 Recognition
	7 Validation and Evaluation
	8 Conclusion
	References
Classification of Vehicle Type on Indian Road Scene Based on Deep Learning
	1 Introduction
	2 Related Work
		2.1 Related Work on Vehicle Classifications
		2.2 Background of Convolution Neural Network
	3 Methodology
		3.1 Foreground and Background Classification
		3.2 Regressor
	4 Experiment and Results
		4.1 Dataset
		4.2 Experimental Results
		4.3 Result Analysis
	5 Conclusion
	References
Indian Road Lanes Detection Based on Regression and clustering using Video Processing Techniques
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Video Acquisition
		3.2 Preprocessing
		3.3 Lane Segment Detection
		3.4 Line Segment Clustering Technique
		3.5 Lane Fitting and Regression
		3.6 Detection of White and Yellow Lines
		3.7 Result and Observations
		3.8 Conclusion and Future Work
	References
Detection of Emotion Intensity Using Face Recognition
	1 Introduction
	2 Related Work
	3 Proposed Design
		3.1 Pre-processing
		3.2 Face Detection
		3.3 Feature Extraction
		3.4 Emotional Intensity Classification and Evaluation
		3.5 Result Analysis
	4 Dataset
	5 Implementation
	6 Results
	7 Conclusion
	8 Future Scope
	References
Double Authentication System Based on Face Identification and Lipreading
	1 Introduction
	2 Literature Survey
	3 Proposed System
		3.1 Face identification
		3.2 Lipreading
	4 Results
	5 Conclusion and Future Scope
	References
Safety Gear Check at Industries and Laboratories Using Convolutional Neural Network Based on Deep Learning
	1 Introduction
	2 Convolutional Neural Networks
	3 Architecture
		3.1 KNN Algorithm
		3.2 Alex NetModel
		3.3 Inception-V3 Model
		3.4 Data Set
	4 Proposed System
	5 Experimental Results
	6 Conclusion
	References
Analysis of Changing Trends in Textual Data Representation
	1 Introduction
	2 Growth of Text Data
		2.1 Word Level Representation
		2.2 Rise of Word Embedding
		2.3 Rise of Sentence/Paragraph/Document Embedding
	3 Comparison between Different Representation Models
		3.1 BoW vs N-gram
		3.2 GloVe vs Word2vec
		3.3 ELMo vs FastText
		3.4 BoW vs Word Embedding (Word2vec or GloVe) Models
		3.5 Comparison Among the Approaches: Traditional Approaches, Word Embedding, and Document Embedding
	4 Conclusion
	References
Detection of Falsary Happening on Social Media Using Image Processing: Feature Extraction and Matching
	1 Introduction
	2 Related Work
		2.1 Scale Invariant Feature Transform (SIFT)
		2.2 Speed Up Robust Feature (SURF)
		2.3 Zernike Moments
		2.4 Hue Moments
	3 Experimental Results
		3.1 Matching Algorithm for MSER Features
		3.2 Matching Algorithm for SURF Features
		3.3 Matching Algorithm for SIFT Features
		3.4 Matching Algorithm with Harris Detector
		3.5 Brute Force Matching
	4 Comparative Analysis
	5 Conclusion
	References
Development of Multi Faces Recognition System Using HOG Features and Neural Network Classifier in Real Time Environment
	1 Introduction
	2 Proposed Face Recognition System
	3 Experimental Result and Analysis
	4 Conclusion
	References
Extraction of Key Frame from Random Videos Based On Discrete Cosine Transformation
	1 Introduction
	2 Related Work
	3 Methodology
	4 Expirimental and Result Analysis
	5 Conclusion
	References
Data Science and Machine Learning
Prediction of SO2 Air Pollution Quality Parameter of Kolhapur City Using Time Series Analysis
	1 Introduction
	2 Review of Literature
	3 Methodology
		3.1 Study Area
		3.2 Time Series Model
	4 Result and Discussion
	5 Conclusion
	References
A Big Data Prediction for Weather Forecast Using Hybrid ARIMA-ANN Time Series Model
	1 Introduction
	2 Review of Literature
	3 Data Source
	4 Research Methodology
		4.1 Hybrid Time Series Model
	5 Empirical Results
	6 Conclusion
	References
Automatic Detection of Riots Using Deep Learning
	1 Introduction
		1.1 Introduction to the Concept
	2 Past Related Work
	3 The Dataset
	4 Video and Camera
	5 Approach
		5.1 Preprocessing
		5.2 Architecture, Explanation and Workflow
		5.3 Experiment
	6 Result
	7 Conclusion and Future Work
	References
A New Method for Defining Scale to Estimate the Aspects Oriented Sentiment Polarity of the Tweets
	1 Introduction
	2 Related Work
	3 Methods and Materials
		3.1 Preprocessing
		3.2 Sentence Level Polarity Estimation (Local Polarity)
		3.3 Aspect Level Local Polarity Ratio
		3.4 Aspect Level Local Polarity Deviation
		3.5 Aspect Level Global Polarity Ratio
		3.6 Aspect Level Global Polarity Deviation
		3.7 Aspect Level Local Probability Scale
		3.8 Aspect Level Global Probability Scale
		3.9 Aspect Level Positive and Negative Probability Scales
		3.10 Estimating Sentiment Polarity
	4 Experimental Study
		4.1 Positive Predictive Value (PPV) & Negative Predictive Value (NPV)
		4.2 True Positive Rate (TPR) & True Negative Rate (TNR)
		4.3 Prediction Accuracy and Harmonic Mean
	5 Conclusion and Future Work
	References
Protecting Big Data Sets from Unauthorized Users on Cloud
	1 Introduction
	2 De-identification
	3 Securing Big Data Storage
	4 Security Issues and Challenges in Big Data
	5 Securing Big Data
	6 Current Cloud Storage Mechanism
	7 Design of Structure for Secure Cloud Storage
	8 Proposed Algorithm
	9 Experimental Results
		9.1 Assessment Execution and Instant Performance
		9.2 Assessment of Overall Upload Time
		9.3 Assessment of Total Download Time
	10 Conclusion
	References
Text Categorization: A Lazy Learning-Based Approach
	1 Introduction
	2 Literature Survey
	3 Proposed Methodology
		3.1 Data
		3.2 Pre-processing
		3.3 Feature Extraction
		3.4 Classification
	4 Result Analysis
		4.1 Comparison with Other Standard Classifiers
		4.2 Comparison with Other Algorithms in Bangla
		4.3 Performance of the System on Other Datasets
	5 Conclusion
	References
Exploring Research Pathways in Record Deduplication and Record Linkage
	1 Introduction
	2 Literature Survey
		2.1 Deterministic Record De-duplication
		2.2 Probabilistic Record Linkage
		2.3 Research Progression in Record De-duplication
	3 Blocking and Indexing Techniques
	4 Adaptive Variants of SNM
	5 Advanced Record De-duplication Techniques
	6 Conclusions
	References
Enhancing Enterprise Business Processes Through AI Based Approach for Entity Extraction – An Overview of an Application
	1 Introduction
	2 Motivation
	3 High Level Design of Framework
	4 About the Dataset
	5 Major Components of the Framework
		5.1 Quality Classifier
		5.2 Optical Character Recognition (OCR)
		5.3 Information Miner
	6 Conclusion
	References
Web Based GIS Village Information System: A Review
	1 Introduction
	2 Literature Review
	3 Village Information System (VIS)
		3.1 Data Can Be Spatial Data, Non-spatial Data, Attribute Data
		3.2 Consolidation of Spatial and Non-spatial Databases
		3.3 The Spatial Database for the Village Can Be Generated by Satellite Images and Can Be Processed by Following Methods
		3.4 Creation of Different Maps for VIS
	4 Applications of VIS
	5 Conclusion
	References
Document Understanding and Recognition
An Approach to Extract the Relation and Location from the Short Stories
	1 Introduction
	2 Literature Review
	3 Proposed Model
		3.1 Text Document (Story)
		3.2 Preprocessing
	4 Experimental Result
		4.1 Actor Extraction
		4.2 Relation Extraction
		4.3 Location Extraction
		4.4 Extracted Event
		4.5 Relation Between Extracted Texts (Feature)
		4.6 Generation of Summary
	5 Experimental Result
	6 Conclusion
	References
Recognition of Partial Handwritten MODI Characters Using Zoning
	1 Introduction
	2 Significance of MODI Script
	3 Comparative Analysis of Character Representations
		3.1 MODI Script-Superset of the Problem
	4 Literature Review for PCR
	5 Proposed Framework for PCR
		5.1 Data Acquisition
		5.2 Pre-processing
		5.3 Extraction PAC
		5.4 Feature Extraction
	6 Evaluation of Zoning Patterns for Degradations
		6.1 Circular Zones
		6.2 Square and Rectangular Zones
		6.3 Irregular Hexagonal Zones
		6.4 Triangular Zones
		6.5 Rhombus Zone
		6.6 Zone Selection
	7 Experiment and Results
		7.1 Distinct and Integrated Approach (Homogeneous Zones)
		7.2 Distinct and Integrated Approach (Heterogeneous Zones)
		7.3 Classification and Recognition
	8 Conclusion
	9 Scope for Further Study
	References
A Modified Approach for the Segmentation of Unconstrained Cursive Modi Touching Characters Cluster
	1 Introduction
	2 Literature Review
	3 System Framework
		3.1 Stage I
		3.2 Stage II
		3.3 Stage III
		3.4 Stage IV
	4 Dataset
	5 Results and Discussions
	6 Conclusions
	References
Resource Creation for Sentiment Analysis of Under-Resourced Language: Marathi
	Abstract
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Corpus-Based Approach
		3.2 SentiWordNet 3.0-Based Approach
		3.3 HSWN-Based Approach
	4 Implementation
		4.1 Corpus-Based Approach
		4.2 SentiWordNet (SWN) 3.0-Based Approach
		4.3 HSWN-Based Approach
	5 Results
		5.1 Corpus-Based Approach
		5.2 SWN-Based Approach
		5.3 HSWN-Based Approach
		5.4 Resource Evaluation
	6 Conclusion
	References
Review on Offline Signature Verification: Datasets, Methods and Challenges
	1 Introduction
	2 Challenges
	3 Datasets
	4 Preprocessing Techniques
	5 Feature Extraction
	6 Trained Model and Performance on Different Datasets
	7 Conclusion
	References
Detection of Fraudulent Alteration of Bank Cheques Using Image Processing Techniques
	1 Introduction
	2 Related Work
	3 Alteration Examples
	4 Proposed Methodology
	5 Proposed Algorithm
	6 Experimental Results
	7 Conclusion
	References
Character Recognition of Offline Handwritten Marathi Documents Written in MODI Script Using Deep Learning Convolutional Neural Network Model
	1 Introduction
	2 State-of-the-Art
	3 Previous Research Work
	4 The Proposed Method
		4.1 Problem Statement
		4.2 Mathematical Formulation
		4.3 Proposed Algorithm
	5 Experimental Results and Discussion
	6 Conclusion and Future Prospects
	References
Recognition of Handwritten Indian Trilingual City Names
	1 Introduction
	2 Related Works on City Name Recognition
	3 Dataset
	4 CNN Based Classification
	5 Result and Discussion
		5.1 Error Analysis and Comparison
	6 Conclusion
	References
Deep Learning for Word-Level Handwritten Indic Script Identification
	1 Introduction
	2 Contribution Outline
		2.1 CNN Architecture
		2.2 Data Representation
	3 Experiments
		3.1 Dataset, and Evaluation Metrics and Protocol
		3.2 Experimental Set up
		3.3 Our Results and Comparative Study
	4 Conclusion
	References
A Survey on Line Segmentation Techniques for Indic Scripts
	1 Introduction
		1.1 Challenges of Line Segmentation
	2 Databases
	3 Overview of Some State-of-the-Art Line Segmentation Techniques
		3.1 Projection Profile Based Methods
		3.2 Hough Transform Based Methods
		3.3 Smearing Based Methods
		3.4 Morphology Based Methods
		3.5 Hybrid and Other Methods
	4 Conclusion and Future Scopes
	References
Peruse and Recognition of Old Kannada Stone Inscription Characters
	1 Introduction
		1.1 Research Objectives
	2 Related Work
	3 Proposed Algorithm
	4 Results
	5 Conclusion
	References
Emotion Recognition Using Standard Deviation and Pitch as a Feature in a Marathi Emotional Utterances
	Abstract
	1 Introduction
	2 Literature Review
	3 Speech Database Used for Emotion Recognition
	4 About Marathi Emotional Speech Database
	5 Feature Extraction
	6 Methodology
	7 Result
	8 Conclusion
	References
Citation Classification Prediction Implying Text Features Using Natural Language Processing and Supervised Machine Learning Algorithms
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Dataset
		3.2 Pre-processing
		3.3 Predictive Models
		3.4 Evaluation Metrices
	4 Results and Discussion
		4.1 Evaluation Results
		4.2 Discussion
	5 Conclusion and Future Work
	References
Author Index