Pattern Recognition: Applications and Methods : 4th International Conference, ICPRAM 2015, Lisbon, Portugal, January 10-12, 2015, Revised Selected Papers / / edited by Ana Fred, Maria De Marsico, Mário Figueiredo

• Pubblicazione: Cham : , : Springer International Publishing : , : Imprint : Springer, , 2015
• Edizione: 1st ed. 2015.
• Descrizione fisica: 1 online resource (XVI, 301 p. 166 illus. in color.)
• Disciplina: 006.4
• Soggetto topico: Pattern recognition; Optical data processing; Artificial intelligence; Computer graphics; Computer simulation; Computational intelligence; Pattern Recognition; Image Processing and Computer Vision; Artificial Intelligence; Computer Graphics; Simulation and Modeling; Computational Intelligence
• Persona (resp. second.): FredAna; De MarsicoMaria; FigueiredoMário
• Note generali: Bibliographic Level Mode of Issuance: Monograph
• Nota di contenuto: Intro -- Preface -- Organization -- Contents -- Invited Paper -- 3D Computer Vision: From Points to Concepts -- 1 Introduction -- 2 3D Sensors -- 3 Keypoints -- 4 Descriptors -- 4.1 Evaluating Descriptors -- 4.2 Genetic Algorithm-Evolved 3D Point Cloud Descriptor -- 5 3D Object Recognition -- 5.1 Set Distances -- 5.2 Deep Transfer Learning for 3D Object Recognition -- 6 3D Object Tracking -- 7 Challenges -- 8 Conclusion -- References -- Theory and Methods -- Density Difference Detection with Application to Exploratory Visualization -- 1 Introduction -- 2 Theoretical Foundations -- 2.1 Scale Space Representation -- 2.2 Kernel Density Estimation -- 2.3 Bandwidth Selection -- 2.4 Multiple Groups -- 3 Detection Framework -- 3.1 Scale Tracing -- 3.2 Spot Detection -- 3.3 Shape Adaption -- 4 Exploratory Visualization -- 4.1 Star Coordinate Visualization -- 4.2 Preserving Orthography -- 4.3 Guiding Explorations -- 5 Experiments -- 5.1 Decision Rule -- 5.2 Data Sets -- 5.3 Parameter Investigation -- 5.4 Classification Results -- 6 Conclusion -- References -- Identifying and Mitigating Labelling Errors in Active Learning -- 1 Introduction -- 2 Background -- 2.1 Active Learning with Uncertainty -- 2.2 Impact of Label Noise -- 3 Characterizing Mislabelled Instances -- 3.1 Mislabelling Likelihood -- 3.2 Informativeness of Possibly Mislabelled Instances -- 4 Mitigating Label Noise -- 4.1 Discard, Weight and Relabel -- 5 Experiments -- 5.1 Mislabelling Likelihood -- 5.2 Label Noise Mitigation -- 6 Conclusion and Future Work -- References -- A Holistic Classification Optimization Framework with Feature Selection, Preprocessing, Manifold Learning and Classifiers -- 1 Introduction -- 2 Automatic Optimization Frameworks -- 3 Classification Pipeline -- 3.1 Feature Selection Element -- 3.2 Feature Preprocessing Element -- 3.3 Feature Transform Element. ; 3.4 Classifier Element -- 3.5 Hyperparameters -- 4 Optimization of the Pipeline Configuration -- 4.1 Extended Evolution Strategies -- 4.2 Optimization Target Function -- 4.3 Multi-pipeline Classifier -- 5 Experiments -- 5.1 Evaluation of the Optimization Process -- 5.2 Analysis of the Optimization Trajectory -- 5.3 Top Configuration Graph -- 5.4 Evaluation of the Generalization -- 6 Conclusions -- References -- Feature Extraction and Learning Using Context Cue and R00E9nyi Entropy Based Mutual Information -- Abstract -- 1 Introduction -- 2 Feature Extraction Using CKD -- 2.1 Formulation of CMK -- 2.2 Approximation of CMK -- 3 Feature Learning Using CSQMI -- 3.1 R00E9nyi Entropy and CSQMI -- 3.2 Codebook Selection and Refinement Using CSQMI -- 3.2.1 Codebook Selection -- 3.2.2 Codebook Refinement -- 4 Experiments -- 4.1 Parameter Configuration -- 4.2 Evaluation of Codebook Learning -- 4.3 Evaluation of Object Recognition -- 4.4 Evaluation of Object Detection -- 5 Conclusion -- Acknowledgements -- References -- Detection of Abrupt Changes in Spatial Relationships in Video Sequences -- 1 Introduction -- 2 Rupture Detection Approach -- 2.1 Fuzzy Object Representation -- 2.2 Spatial Relationships and Rupture Detection -- 3 Experiments and Evaluations -- 3.1 Parameters Tuning -- 3.2 Ruptures in Spatial Relationships -- 3.3 Impact of Object Representation -- 4 Conclusion -- References -- Diffusion-Based Similarity for Image Analysis -- 1 Introduction -- 2 Diffusion Distance and Clustering -- 3 The Problems of Diffusion Distance -- 4 Normalised Diffusion Cosine Similarity -- 5 Experimental Results -- 6 Conclusions -- References -- Automatic Detection and Recognition of Symbols and Text on the Road Surface -- 1 Introduction -- 2 Related Work -- 3 Detection and Sorting of Candidate Regions -- 3.1 Detection of Candidate Regions. ; 3.2 Sorting of Candidates into Words and Symbols -- 4 Recognition of Words -- 4.1 Correction of Perspective Distortion of Words -- 4.2 OCR of Region -- 5 Recognition of Symbols -- 5.1 Selection of Classifier -- 5.2 Selection of HOG Block Density -- 6 Temporal Information -- 6.1 Calculation of Road Movement -- 6.2 Matching and Temporal Fusion of Words and Symbols -- 7 Results -- 7.1 Results for Text Detection and Recognition -- 7.2 Results for Symbol Detection and Recognition -- 7.3 Failure Cases -- 8 Conclusion -- References -- Applications -- Using BLSTM for Spotting Regular Expressions in Handwritten Documents -- 1 Introduction -- 2 Related Work -- 3 BLSTM-CTC/HMM System -- 3.1 Character Recognition and Segmentation -- 3.2 Handwritten Word Spotting Model -- 3.3 Regular Expression Spotting Model -- 4 Experiments -- 4.1 Features Set -- 4.2 Results and Discussion -- 4.3 Using the BLSTM Without HMM -- 5 Conclusion -- References -- A Similarity-Based Color Descriptor for Face Detection -- 1 Introduction -- 2 Three-Patch LBP Descriptor and a Multi Scale Variant -- 3 A New Color Descriptor - Coupled-Chroma TPLBP -- 4 Evaluation of Color Descriptors -- 4.1 Invariance to Photometric and Geometric Variations -- 4.2 Invariance to Gaussian Noise -- 4.3 Discriminative Power -- 5 Evaluation of Color Descriptors in a Face Detection Setting -- 5.1 Dataset -- 5.2 Evaluation Protocol -- 5.3 Results -- 6 Conclusions -- References -- Pose Estimation and Movement Detection for Mobility Assessment of Elderly People in an Ambient Assisted Living Application -- 1 Introduction -- 2 Related Work -- 3 Mobility Assessment -- 3.1 Person Localisation -- 3.2 Movement Detection -- 3.3 Pose Estimation -- 3.4 Human Machine Interface -- 4 Experimental Results -- 5 Action Recognition -- 6 Conclusions -- 7 Future Work -- References. ; A Non-rigid Face Tracking Method for Wide Rotation Using Synthetic Data -- 1 Introduction -- 2 Face Representation -- 2.1 Shape Representation -- 2.2 Projection -- 2.3 Appearance Representation -- 3 Our Method -- 3.1 Model Training from Synthesized Data -- 3.2 Robust Initialization -- 3.3 Matching Strategy by Keyframes -- 3.4 Fitting via Pose-Wise Classifiers -- 4 Experimental Results -- 5 Conclusions -- References -- 3-D Face Recognition Using Geodesic-Map Representation and Statistical Shape Modelling -- 1 Introduction -- 2 Sparse Facial Landmark Detection -- 3 Geodesic-Map Representation -- 4 Geodesic-Map Matching -- 5 Dimensionality Reduction -- 6 New Dataset Fitting -- 7 Experimental Results -- 7.1 Facial Expression Changes -- 7.2 Data Resolution Variation -- 7.3 Missing Data -- 8 Conclusions -- References -- Learning Discriminative Mid-Level Patches for Fast Scene Classification -- 1 Introduction -- 2 Related Work -- 3 Discovering Discriminative Patches: Designed for Speed -- 3.1 Patch Selection and Feature Extraction -- 3.2 Classifier Training -- 3.3 Classifier Selection -- 4 Image Representation and Scene Classification -- 5 Experiments and Results -- 6 Conclusion -- References -- Modification of Polyp Size and Shape from Two Endoscope Images Using RBF Neural Network -- 1 Introduction -- 2 VBW Model -- 3 Proposed Approach -- 3.1 Estimating Reflectance Parameter -- 3.2 NN Learning for Modification of Surface Gradient -- 3.3 NN Generalization and Modification of Z -- 3.4 Regression Analysis for Shape Modification -- 4 Experimental Results -- 4.1 NN Learning -- 4.2 Order Number in Polynomial with Regression Analysis -- 4.3 Computer Simulation -- 4.4 Real Image Experiments -- 5 Conclusion -- References -- Detecting and Dismantling Composite Visualizations in the Scientific Literature -- 1 Introduction -- 2 Related Work. ; 3 Decomposition Algorithm -- 3.1 Step 1: Splitting -- 3.2 Step 2: Merging -- 3.3 Step 3: Selecting -- 4 Composite Figure Detection -- 5 Experimental Evaluation -- 6 Limitations and Future Work -- 7 Conclusions -- References -- Tensor Deep Stacking Networks and Kernel Deep Convex Networks for Annotating Natural Scene Images -- 1 Introduction -- 2 Convex Deep Learning Models -- 2.1 Tensor Deep Stacking Networks -- 2.2 Kernel Deep Convex Networks -- 2.3 Framework for Image Annotation -- 3 Experiments and Results -- 3.1 Experimental Setup -- 3.2 Feature Extraction -- 3.3 Datasets Used -- 3.4 Results -- 4 Summary and Conclusions -- References -- MOSAIC: Multi-object Segmentation for Assisted Image ReConstruction -- 1 Introduction -- 2 MOSAIC Procedures -- 2.1 Fragment Catalogue -- 2.2 Fragment Search -- 3 MOSAIC Interface -- 4 Experimental Results -- 4.1 An Experiment on a Real Life Case -- 5 Conclusions -- References -- Author Index.
• Sommario/riassunto: This book constitutes the thoroughly refereed post-conference proceedings of the 4th International Conference on Pattern Recognition, ICPRAM 2015, held in Lisbon, Portugal, in January 2015. The 20 revised full papers were carefully reviewed and selected from 145 submissions and describe up-to-date applications of pattern recognition techniques to real-world problems, interdisciplinary research, experimental and/or theoretical studies yielding new insights that advance pattern recognition methods.
• Titolo autorizzato: Pattern Recognition Applications and Methods
• ISBN: 3-319-27677-8
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910483798303321
• Lo trovi qui: Univ. Federico II
• Serie: Image Processing, Computer Vision, Pattern Recognition, and Graphics ; ; 9493