Combining rules using local statistics and uncertainty estimates for improved ensemble segmentation

Segmentation using an ensemble of classifiers (or committee machine) combines multiple classifiers’ results to increase the performance when compared to single classifiers. In this paper, we propose new concepts for combining rules. They are based (1) on uncertainties of the individual classifiers,...

Full description

Saved in:
Bibliographic Details
Published inPattern recognition and image analysis Vol. 27; no. 3; pp. 444 - 457
Main Authors Al-Taie, A., Hahn, H. K., Linsen, L.
Format Journal Article
LanguageEnglish
Published Moscow Pleiades Publishing 01.07.2017
Springer Nature B.V
Subjects
Analysis
Classifiers
Computer Science
Image Processing and Computer Vision
Noise reduction
Pattern Recognition
Performance enhancement
Processing
Representation
Segmentation
Uncertainty
Understanding of Images
modified Fuzzy c-means
combining rules
image segmentation
ensemble of classifiers
Online AccessGet full text

Cover

More Information
Summary:Segmentation using an ensemble of classifiers (or committee machine) combines multiple classifiers’ results to increase the performance when compared to single classifiers. In this paper, we propose new concepts for combining rules. They are based (1) on uncertainties of the individual classifiers, (2) on combining the result of existing combining rules, (3) on combining local class probabilities with the existing segmentation probabilities at each individual segmentation, and (4) on using uncertainty-based weights for the weighted majority rule. The results show that the proposed local-statistics-aware combining rules can reduce the effect of noise in the individual segmentation result and consequently improve the performance of the final (combined) segmentation. Also, combining existing combining rules and using the proposed uncertainty- based weights can further improve the performance.
ISSN:1054-6618
1555-6212
DOI:10.1134/S105466181703004X