Automatic parameter selection for a k-segments algorithm for computing principal curves

Authors:
Haonan Wang;Thomas C. M. Lee
Affiliations:
Department of Statistics, Colorado State University, Fort Collins, CO 80523-1877, USA;Department of Statistics, Colorado State University, Fort Collins, CO 80523-1877, USA
Venue:
Pattern Recognition Letters
Year:
2006

Citing 8
Cited 2

Growing cell structures—a self-organizing network for unsupervised and supervised learning

Neural Networks
GTM: the generative topographic mapping

Neural Computation
Learning and Design of Principal Curves

IEEE Transactions on Pattern Analysis and Machine Intelligence
Stochastic Complexity in Statistical Inquiry Theory

Stochastic Complexity in Statistical Inquiry Theory
A k-segments algorithm for finding principal curves

Pattern Recognition Letters
A robust algorithm for image principal curve detection

Pattern Recognition Letters
Wavelet thresholding via MDL for natural images

IEEE Transactions on Information Theory
Spatially adaptive wavelet denoising using the minimum description length principle

IEEE Transactions on Image Processing

Extraction of curvilinear features from noisy point patterns using principal curves

Pattern Recognition Letters
Regularization-free principal curve estimation

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.10

Visualization

Abstract

This paper studies the k-segments algorithm proposed by Verbeek et al. [Verbeek, J.J., Vlassis, N., Krose, B., 2002. A k-segments algorithm for finding principal curves, Pattern Recognition Lett. 23, 1009-1017] for computing principal curves. In particular an automatic method for choosing the ''free'' parameters in this k-segments algorithm is proposed. Experimental results are provided to demonstrate the performance of this proposed method.