Unsupervised segmentation and approximation of digital curves with rate-distortion curve modeling

This paper considers the problem of unsupervised segmentation and approximation of digital curves and trajectories with a set of geometrical primitives (model functions). An algorithm is proposed based on a parameterized model of the Rate–Distortion curve. The multiplicative cost function is then de...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Pattern recognition Jg. 47; H. 2; S. 623 - 633
Hauptverfasser: Kolesnikov, Alexander, Kauranne, Tuomo
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Kidlington Elsevier Ltd 01.02.2014
Elsevier
Schlagworte:
Algorithms
Applied sciences
Approximation
Cost function
Curve fitting
Digital
Exact sciences and technology
Graphical model
Information, signal and communications theory
Mathematical analysis
Mathematical models
Piecewise linear approximation
Segmentation
Segments
Shape
Signal and communications theory
Signal representation. Spectral analysis
Signal, noise
Telecommunications and information theory
Graphical model
Shape
Piecewise linear approximation
Curve fitting
Line segment
Segmentation
Time series
Rate distortion theory
Unsupervised classification
Algorithm
Signal classification
Modeling
Polygonal approximation
Geometrical model
Algorithm performance
Heuristic method
Approximation error
Cost function
Piecewise linearization
ISSN:0031-3203, 1873-5142
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper considers the problem of unsupervised segmentation and approximation of digital curves and trajectories with a set of geometrical primitives (model functions). An algorithm is proposed based on a parameterized model of the Rate–Distortion curve. The multiplicative cost function is then derived from the model. By analyzing the minimum of the cost function, a solution is defined that produces the best possible balance between the number of segments and the approximation error. The proposed algorithm was tested for polygonal approximation and multi-model approximation (circular arcs and line segments for digital curves, and polynomials for trajectory). The algorithm demonstrated its efficiency in comparisons with known methods with a heuristic cost function. The proposed method can additionally be used for segmentation and approximation of signals and time series. •A new algorithm for unsupervised segmentation of digital curves is introduced.•This method gives solutions with the best balance between error and description length.•A multiplicative criterion for evaluation of solutions is introduced.
Bibliographie:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0031-3203
1873-5142
DOI:10.1016/j.patcog.2013.09.002