Animal Biometrics : Techniques and Applications

"This book presents state-of-the-art methodologies and a comprehensive introduction to the recognition and representation of species and individual animals based on their physiological and phenotypic appearances, biometric characteristics, and morphological image patterns. It provides in-depth coverage of this emerging area, with an emphasis on the design and analysis techniques used in visual animal biometrics-based recognition systems. The book offers a comprehensive introduction to visual animal biometrics, addressing a range of recent advances and practices like sensing, feature extraction, feature selection and representation, matching, indexing of feature sets, and animal biometrics-based multimodal systems. It provides authoritative information on all the major concepts, as well as highly specific topics, e.g. the identification of cattle based on their muzzle point image pattern and face images to prevent false insurance claims, or the monitoring and registration of animals based on their biometric features. As such, the book provides a sound platform for understanding the Visual Animal Biometrics paradigm, a vital catalyst for researchers in the field, and a valuable guide for professionals. In addition, it can help both private and public organizations adapt and enhance their classical animal recognition systems."--Back cover Preface 5 Special Acknowledgements 9 Contents 10 About the Authors 15 Abbreviations 17 List of Figures 19 List of Tables 24 1 Animal Biometrics: Concepts and Recent Application 26 Abstract 26 1.1 Introduction 26 1.2 Interdisciplinary Relevance of Animal Biometrics 29 1.3 Promising Applications of Animal Biometrics-Based Recognition Systems 30 1.4 Prerequisites for Promising Applications of Animal Biometrics 32 1.5 Animal Biometrics Recognition System 33 1.6 Major Component of Animal Biometrics-Based Recognition System 34 1.6.1 Data Acquisition and Data Preprocessing 36 1.6.2 Extraction and Representation of Features 36 1.6.3 Matching of Animal Biometric Features 36 1.7 Current State-of-the-Art Animal Biometrics Recognition Systems 37 1.7.1 SLOOP: A Pattern Recognition-Based System for Animal Recognition 37 1.7.2 Animal Biometrics-Based Recognition System for Face Detection of Chimpanzee 39 1.7.3 ECOCEAN Whale Shark Identification System 40 1.8 Issues and Challenges of Animal Biometrics 41 1.8.1 Identification of Animal Based on Coat Pattern 41 1.8.2 Segmentation of Background and Discriminatory Features in Natural Environments 41 1.9 Summary 43 References 43 2 Analytical Study of Animal Biometrics: A Technical Survey 46 Abstract 46 2.1 Introduction 46 2.2 Classical Animal Identification Methodology 48 2.2.1 Permanent Identification Methodology 49 2.2.1.1 Ear-Notching-Based Cattle Identification 49 2.2.1.2 Ear-Tattooing-Based Marking Method for Animal 51 2.2.1.3 Hot iron-Branding-Based Marking Method for Animal 51 2.2.1.4 Freeze-Branding-Based Animal Marking Method for Animal 52 2.2.2 Semipermanent Animal Identification Methodology 52 2.2.2.1 Collar-Based Marking Method for Cattle Identification 52 2.2.2.2 Ear-Tagging-Based Animal Identification Method 53 2.2.3 Temporary Animal Identification Methodology 53 2.2.3.1 RFID-Based Identification Method of Animals 54 2.2.3.2 Sketch Pattern-Based Animal Identification 54 2.3 Visual Animal Biometrics: Current Trends of Animal Recognition Methodologies 57 2.3.1 Retinal Vascular Pattern-Based Cattle Identification 57 2.3.2 Iris Pattern-Based Cattle Identification 59 2.3.3 Cattle Identification Using Biometric Features 60 2.3.4 Muzzle Print Image-Based Cattle Identification 60 2.3.5 Muzzle Point Image Pattern-Based Cattle Identification 64 2.3.6 Identification of Cattle Based on Face Biometric Feature 67 2.4 Morphological Image Pattern-Based Animal Identification 68 2.4.1 Identification and Classification of Animal Using Profiling Behavior 71 2.5 Research Contribution, Communities, Simulation Tools, and Sharing of Database 83 2.6 Current Trends and Methods 85 2.7 Technical Issues and Challenges 88 2.7.1 Algorithmic Challenges 88 2.7.2 Integration and Unification Structure of Output Data 90 2.7.3 Feature Extraction and Representation Challenges 90 2.7.4 Challenges of Feature Extraction and Representation of Species Using Holistic and Texture Feature-Based Techniques 91 2.8 Summary 94 References 94 3 Recognition of Cattle Using Face Images 104 Abstract 104 3.1 Introduction 104 3.2 Motivation Behind the Work 105 3.3 Preparation and Description of Face Database 107 3.4 Proposed Cattle Recognition System 111 3.4.1 Sensor Module (Data Acquisition Phase) 112 3.4.2 Preprocessing and Enhancement Phase 113 3.4.2.1 Algorithm of Face Recognition of Cattle 113 Keeping Components and Generating the Feature Vector 114 Testing of the New Face Dataset of Cattle 114 3.4.3 Challenges of Face Recognition in Cattle 114 3.5 Feature Extraction and Matching Phase 117 3.6 Experimental Result and Discussion 119 3.6.1 Performance Evaluation 119 3.7 Experimental Protocol and Analysis 121 3.8 Summary 130 References 132 4 Muzzle Point Pattern-Based Techniques for Individual Cattle Identification 136 Abstract 136 4.1 Introduction 136 4.1.1 Motivation Behind the Work 138 4.1.2 Major Contributions of the Research Work 139 4.2 Biometric Characteristics of Muzzle Point Images 140 4.3 Proposed System 140 4.3.1 Preprocessing and Enhanced of Muzzle Point Image Pattern 142 4.3.2 Image Enhancement Using CLAHE Technique 142 4.3.3 Segmentation and Feature Extraction 145 4.3.4 Matching of Muzzle Point Image Using Chi-Square-Based Matching Technique 146 4.4 Experimental Results and Discussion 148 4.4.1 Database Preparation and Description 148 4.4.2 Performance Evaluations of Proposed Algorithm 150 4.4.2.1 Appearance-Based Feature Extraction and Representation Technique 151 4.4.2.2 Texture Feature-Based Descriptor Algorithms 152 4.4.3 Experimental Evaluation 153 4.5 Summary and Future Directions 156 References 157 5 Identification of Cattle Based on Muzzle Point Pattern: A Hybrid Feature Extraction Paradigm 161 Abstract 161 5.1 Introduction 161 5.1.1 Major Contribution of the Work 164 5.2 Materials and Methods 164 5.3 Proposed Cattle Recognition System 165 5.3.1 Preprocessing and Enhancement Process of Muzzle Point of Cattle 165 5.3.2 Segmentation of Muzzle Point Image 167 5.3.2.1 Texture-Based Segmentation and Color-Based Segmentation 168 5.3.2.2 Minima Selection and Region Merging Using Watershed Segmentation Technique 170 5.4 Feature Extraction and Matching 171 5.5 Experimental Results Performance Evaluation 172 5.6 Performance Analysis 175 5.7 Recognition of Muzzle Point Image Under Different Rotations 178 5.8 Recognition of Muzzle Image Pattern Under Different Occlusion Conditions 178 5.9 Summary 180 References 182 6 Deep Learning Framework for Recognition of Cattle Using Muzzle Point Image Pattern 186 Abstract 186 6.1 Introduction 186 6.1.1 Motivation Behind the Work 187 6.1.2 Major Research Contributions 188 6.2 Biometric Characteristics of Muzzle Point Image Pattern of Cattle 189 6.3 Proposed System 190 6.3.1 Preprocessing and Enhancement 190 6.3.2 Feature Extraction and Representation 191 6.3.2.1 Convolution Neural Network-Based Recognition Framework 193 6.3.2.2 Classification of Features Using Deep Belief Network and Restricted Boltzmann Machines 195 6.4 Stacked Denoising Auto-encoder Technique 197 6.5 Pretraining and Generalizability of Proposed Recognition Model 200 6.6 Experimental Results and Discussions 201 6.6.1 Performance Evaluations 201 6.6.2 Comparative Analysis 210 6.7 Summary 214 References 215 7 Real-Time Recognition of Cattle Using Fisher Locality Preserving Projection Method 219 Abstract 219 7.1 Introduction 219 7.1.1 Contribution Are Illustrated in Brief 221 7.2 Real-Time Cattle Recognition System 222 7.3 Proposed FLPP Feature Extraction and Representation Approach 226 7.4 Feature Encoding 229 7.4.1 Classification Model Using One-Shot Similarity and One-Class-Based SVM 230 7.4.2 Fishers Linear Discriminant Analysis (FLDA) and OSS Classification Techniques 231 7.5 Matching of Muzzle Point Images 232 7.6 Algorithm for Evaluation for Experimental Results 234 7.6.1 Appearance-Based and Texture-Based Feature Extraction Algorithm 234 7.7 Experimental Result and Analysis 235 7.7.1 Performance Evaluation 235 7.8 Summary 240 References 240 8 Biometric Methods for Animal: Recent Trends and Future Challenges 244 Abstract 244 8.1 Introduction 244 8.2 Animal Biometrics-Based Recognition Systems 246 8.2.1 Low-Cost Cattle Recognition System Using Multimedia Wireless Network 247 8.2.2 Cattle Verification System Using Smart Devices 247 8.3 Identification and Monitoring of Pet Animal Using Animal Biometrics 248 8.4 Horse Identification System Using Animal Biometrics 251 8.5 Face Recognition Framework for Chimpanzee Using Deep Learning Approach 253 8.6 Shared Tools, Availability of Animal Database 255 8.7 Visual Animal Biometrics: Issues and Challenges 255 8.8 Summary and Future Directions 260 References 261 Front Matter ....Pages i-xxviii Animal Biometrics: Concepts and Recent Application (Santosh Kumar, Sanjay Kumar Singh, Rishav Singh, Amit Kumar Singh)....Pages 1-20 Analytical Study of Animal Biometrics: A Technical Survey (Santosh Kumar, Sanjay Kumar Singh, Rishav Singh, Amit Kumar Singh)....Pages 21-78 Recognition of Cattle Using Face Images (Santosh Kumar, Sanjay Kumar Singh, Rishav Singh, Amit Kumar Singh)....Pages 79-110 Muzzle Point Pattern-Based Techniques for Individual Cattle Identification (Santosh Kumar, Sanjay Kumar Singh, Rishav Singh, Amit Kumar Singh)....Pages 111-135 Identification of Cattle Based on Muzzle Point Pattern: A Hybrid Feature Extraction Paradigm (Santosh Kumar, Sanjay Kumar Singh, Rishav Singh, Amit Kumar Singh)....Pages 137-161 Deep Learning Framework for Recognition of Cattle Using Muzzle Point Image Pattern (Santosh Kumar, Sanjay Kumar Singh, Rishav Singh, Amit Kumar Singh)....Pages 163-195 Real-Time Recognition of Cattle Using Fisher Locality Preserving Projection Method (Santosh Kumar, Sanjay Kumar Singh, Rishav Singh, Amit Kumar Singh)....Pages 197-221 Biometric Methods for Animal: Recent Trends and Future Challenges (Santosh Kumar, Sanjay Kumar Singh, Rishav Singh, Amit Kumar Singh)....Pages 223-243