The Science of Pattern Recognition. Achievements and Perspectives

  title={The Science of Pattern Recognition. Achievements and Perspectives},
  author={Robert P. W. Duin and Elzbieta Pekalska},
  booktitle={Challenges for Computational Intelligence},
Automatic pattern recognition is usually considered as an engineering area which focusses on the development and evaluation of systems that imitate or assist humans in their ability of recognizing patterns. It may, however, also be considered as a science that studies the faculty of human beings (and possibly other biological systems) to discover, distinguish, characterize patterns in their environment and accordingly identify new observations. The engineering approach to pattern recognition is… 
Philosophical aspects in pattern recognition research
Pattern recognition is the discipline which studies theories and methods to build machines that are able to discover regularities in noisy data. As many of its characterizations suggest, the field is
Pattern recognition between science and engineering: A red herring?
Structural Inference of Sensor-Based Measurements
The problems and possibilities of general structural inference approaches for the family of sensor-based measurements: images, spectra and time signals, assuming a continuity between measurement samples are summarized.
Systems of neuron image recognition for solving problems of automated diagnoses of neurodegenerative diseases
In the survey, an analysis of the methods and systems meant for automated neuron image analysis based on studies published in leading scientific journals within the past 25 years is presented.
Parsimonious design of pattern recognition systems for slope stability analysis
This paper shows that two particular problems of slope stability prediction can be successfully solved by much more simpler pattern recognition methods, and emphasizes on the importance of data visualization and incremental evaluation during the design cycle of a parsimonious pattern recognition system.
How Mature Is the Field of Machine Learning?
It is proposed to address the question whether the fields of machine learning and pattern recognition have achieved the level of maturity in the sense suggested by Thomas Kuhn, and to discuss the philosophical underpinnings of much of contemporary machine learning research.
A Comparison between Time-Frequency and Cepstral Feature Representations for the Classification of Seismic-Volcanic Signals
The most common representations that have been applied in the literature on classification of seismic-volcanic signals; namely, time-frequency features and cepstral coefficients are described and discussed.
A Novel Hierarchical Object Recognition Algorithm Based on Saliency Analysis
  • Yueming HuYi Li
  • Computer Science
    2018 37th Chinese Control Conference (CCC)
  • 2018
A fast hierarchical algorithm based on saliency analysis for 3D-object recognition is proposed to eliminate the large and complex computation of the present pattern recognition system based on RGB-D
Structural generative descriptions for temporal data
Two novel domain-independent representation frameworks for temporal data suitable for off-line and online mining tasks are formulated and a novel framework for multidimensional data stream evolution diagnosis incorporating RWDE into the context of Velocity Density Estimation (VDE) is formulated.


Open Issues in Pattern Recognition
Old and new open issues are discussed that have to be faced in advancing real world applications that may only be overcome by brute force procedures, while others may be solved or circumvented either by novel and better procedures, or by a better understanding of their causes.
Statistical Pattern Recognition: A Review
The objective of this review paper is to summarize and compare some of the well-known methods used in various stages of a pattern recognition system and identify research topics and applications which are at the forefront of this exciting and challenging field.
Four Scientific Approaches to Pattern Recognition
Four scientific approaches to pattern recognition may be distinguished, explained here and illustrated by some examples.
Statistical Pattern Recognition
  • J. Davis
  • Computer Science
  • 2003
This chapter introduces the subject of statistical pattern recognition (SPR). It starts by considering how features are defined and emphasizes that the nearest neighbor algorithm achieves error rates
Why Classical Models for Pattern Recognition are Not Pattern Recognition Models
A simple explanation of why the classical, or vector-space-based, models for pattern recognition are fundamentally inadequate as such is outlined, based on a radically new understanding of the nature of inductive learning processes.
The science of "pattern recognition".
The unique task of the clinician is primarily a goal-directed intellectual process involving decision-making leading to courses of action, which defies easy mensuration.
A note on core research issues for statistical pattern recognition
The dissimilarity representation , a basis for a domain-based pattern recognition ?
The dissimilarity representation aims at treating objects in their wholeness, avoiding the use of isolated features, and proper knowledge of class densities is not needed, which opens the possibility to a domain based classification in which the training set should be just representative for the domain of the classes.
Introduction to Statistical Pattern Recognition
Two approaches to dimensionality reduction, namely feature selection (FS) and feature extraction (FE) are specified, though FS is a special case of FE, they are very different from a practical viewpoint and thus must be considered separately.
39 Dimensionality and sample size considerations in pattern recognition practice