Undesirable effects of output normalization in multiple classifier systems

2003-06-01
Altincay, H
Demirekler, Mübeccel
Incomparability of the classifier output scores is a major problem in the combination of different classification systems. In order to deal with this problem, the measurement level classifier outputs are generally normalized. However, empirical results have shown that output normalization may lead to some undesirable effects. This paper presents analyses for some most frequently used normalization methods and it is shown that the main reason for these undesirable effects of output normalization is the dimensionality reduction in the output space. An artificial classifier combination example and a real-data experiment are provided where these effects are further clarified.
PATTERN RECOGNITION LETTERS

Suggestions

On output independence and complementariness in rank-based multiple classifier decision systems
Saranlı, Afşar (Elsevier BV, 2001-12-01)
This study presents a theoretical analysis of output independence and complementariness between classifiers in a rank-based multiple classifier decision system in the context of the partitioned observation space theory. To enable such an analysis, an information theoretic interpretation of a rank-based multiple classifier system is developed and basic concepts from information theory are applied to develop measures for output independence and complementariness. It is shown that output independence of classi...
A statistical approach to sparse multi-scale phase-based stereo
Ulusoy, İlkay (Elsevier BV, 2007-09-01)
In this study, a multi-scale phase based sparse disparity algorithm and a probabilistic model for matching uncertain phase are proposed. The features used are oriented edges extracted using steerable filters. Feature correspondences are estimated using phase-similarity at multiple scale using a magnitude weighting scheme. In order to achieve sub-pixel accuracy in disparity, we use a fine tuning procedure which employs the phase difference between corresponding feature points. We also derive a probabilistic ...
Continuous dimensionality characterization of image structures
Felsberg, Michael; Kalkan, Sinan; Kruger, Norbert (Elsevier BV, 2009-05-04)
Intrinsic dimensionality is a concept introduced by statistics and later used in image processing to measure the dimensionality of a data set. In this paper, we introduce a continuous representation of the intrinsic dimension of an image patch in terms of its local spectrum or, equivalently, its gradient field. By making use of a cone structure and barycentric co-ordinates, we can associate three confidences to the three different ideal cases of intrinsic dimensions corresponding to homogeneous image patche...
Resampling approach for cluster model selection
Volkovich, Z.; Barzily, Z.; Weber, Gerhard Wilhelm; Toledano-Kitai, D.; Avros, R. (Springer Science and Business Media LLC, 2011-10-01)
In cluster analysis, selecting the number of clusters is an "ill-posed" problem of crucial importance. In this paper we propose a re-sampling method for assessing cluster stability. Our model suggests that samples' occurrences in clusters can be considered as realizations of the same random variable in the case of the "true" number of clusters. Thus, similarity between different cluster solutions is measured by means of compound and simple probability metrics. Compound criteria result in validation rules em...
Camera auto-calibration using a sequence of 2D images with small rotations
Hassanpour, R; Atalay, Mehmet Volkan (Elsevier BV, 2004-07-02)
In this study, we describe an auto-calibration algorithm with fixed but unknown camera parameters. We have modified Triggs' algorithm to incorporate known aspect ratio and skew values to make it applicable for small rotation around a single axis. The algorithm despite being a quadratic one is easy to solve. We have applied the algorithm to some artificial objects with known size and dimensions for evaluation purposes. In addition, the accuracy of the algorithm has been verified using synthetic data. The des...
Citation Formats
H. Altincay and M. Demirekler, “Undesirable effects of output normalization in multiple classifier systems,” PATTERN RECOGNITION LETTERS, pp. 1163–1170, 2003, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/57537.