Computational Intelligence Paradigms in Advanced Pattern Classification (Studies in Computational Intelligence, 386)

✍ Scribed by Marek R. Ogiela (editor), Lakhmi C Jain (editor)

• Publisher: Springer
• Year: 2012
• Tongue: English
• Leaves: 206
• Category: Library

✦ Synopsis

This monograph presents selected areas of application of pattern recognition and classification approaches including handwriting recognition, medical image analysis and interpretation, development of cognitive systems for image computer understanding, moving object detection, advanced image filtration and intelligent multi-object labelling and classification. It is directed to the scientists, application engineers, professors, professors and students will find this book useful.

✦ Table of Contents

Title
Preface
Contents
Recent Advances in Pattern Classification
New Directions in Pattern Classification
References
Neural Networks for Handwriting Recognition
Introduction
State-of-the-Art
Contribution
Data Processing
General Processing Steps
Our Online System
Our Offline System
Neural Network Based Recognition
Recurrent Neural Networks (RNNs)
Long Short-Term Memory (LSTM)
Bidirectional Recurrent Neural Networks
Connectionist Temporal Classification (CTC)
Multidimensional Recurrent Neural Networks
Hierarchical Subsampling Recurrent Neural Networks
Experiments
Comparison with HMMs on the IAM Databases
Recognition Performance of MDLSTM on Contest’ Data
Conclusion
References
Moving Object Detection from Mobile
Platforms Using Stereo Data Registration
Introduction
Related Work
Proposed Approach
System Setup
Feature Detection and Tracking
Robust Registration
Frame Subtraction
Experimental Results
Conclusions
References
Pattern Classifications in Cognitive Informatics
Introduction
Semantic Analysis Stages
Semantic Analysis vs. Cognitive Informatics
Example of a Cognitive UBIAS System
Conclusions
References
Optimal Differential Filter on Hexagonal Lattice
Introduction
Preliminaries
Least Inconsistent Image
Point Spread Function
Condition for Gradient Filter
Numerical Optimization
Theoretical Evaluation
Signal-to-Noise Ratio
Localization
Experimental Evaluation
Construction of Artificial Images
Detection of Gradient Intensity and Orientation
Overington’s Method of Orientation Detection
Relationship between Derived Filter and Staunton Filter
Experiment and Results
Discussion
Summary
References
Graph Image Language TechniquesSupporting Advanced Classification and
Cognitive Interpretation of CT Coronary Vessel Visualizations
Introduction
The Classification Problem
Stages in the Analysis of CT Images under a Structural
Approach Utilising Graph Techniques
Parsing Languages Generated by Graph Grammars
Picture Grammars in Classification and Semantic
Interpretation of 3D Coronary Vessels Visualisations
Characteristics of the Image Data
Preliminary Analysis of 3D Coronary Vascularisation
Reconstructions
Graph-Based Linguistic Formalisms in Spatial Modelling of
Coronary Vessels
Detecting Lesions and Constructing the Syntactic Analyser
Selected Results
Conclusions Concerning the Advanced Classification and
Cognitive Interpretation of CT Coronary Vessel Visualizations
References
A Graph Matching Approach to Symmetry
Detection and Analysis
Introduction
Symmetries and Their Properties
Rotational Symmetry
Reflectional Symmetry
Interrelations between Rotational and ReflectionalSymmetries
Discussion
Previous Work
Previous Work in Symmetry Detection and Analysis
Local Features
Spectral Matching of Sets of Points in Rn
Spectral Symmetry Analysis
Spectral Symmetry Analysis of Sets in $R$n
Spectral Symmetry Analysis of Images
Experimental Results
Symmetry Analysis of Images
Statistical Accuracy Analysis
Analysis of Three-Dimensional Symmetry
Implementation Issues
Additional Results
Conclusions
References
Pattern Classification Methods for Analysis and
Visualization of Brain Perfusion CT Maps
Introduction
Interpretation of Perfusion Maps – Long and Short Time
Prognosis
Image Processing and Abnormality Detection
Image Registration
Affine Registration
FFD Registration
Thirion’s Demons Algorithm
Comparison of Registration Algorithms
Classification of Detected Abnormalities
System Validation and Results
Data Visualization – Augmented Reality Environment
Augmented Reality Environment
Real Time Rendering of 3D Data
Augmented Desktop - System Performance Test
Summary
References
Inference of Co-occurring Classes: Multi-class
and Multi-label Classification
Introduction
Applications
The Classification Process
Data and Annotation
Classification Approaches
Binary Classification
Multi-class Classification
Multi-label Classification
Multi-class Classification
Multiple Binary classifiers
Direct Multi-class Classification
Associative Classification
Multi-label Classification
Semi-supervised (Annotation) Methods
Inference of Co-occurring Affective States from Non-verbal
Speech
Summary
References
Author Index