%0 Journal Article %1 Barlach1990 %A Barlach, Flemming %C New York, NY, USA %D 1990 %I Elsevier Science Inc. %J Pattern Recognition (PR) %K Coin Decoupled Optimized classification; coins, covariance discriminators linear matrix; %N 3--4 %P 379--384 %R http://dx.doi.org/10.1016/0031-3203(90)90024-F doi:10.1016/j.physletb.2003.10.071 %T Payphone coin validation using pattern recognition %U http://www.sciencedirect.com/science/article/B6V14-48MPKH1-T1/2/f60fddbb8b5ad144210229e877f1ceac %V 23 %X In this paper we give a classifier design approach, which yields classifiers of high computational efficiency and low memory requirements. The method is based on subdivision of the pattern space by a minimum number of optimized linear inequalities, i.e. the discriminators, into regions containing the pattern classes. The discrimination vectors are selected from the principal axes of each pattern class covariance matrix. We present an optimization procedure for sets of discrete parameter bounds. We briefly discuss the problem of multimodal pattern classes, and the possibility of using suboptimum solutions. The main objective is to design a group of optimum classifiers for payphone coin classification. In two examples we compare our classifier performance with the classification performance derived from fitting normal probability densities to the pattern classes.

@article{Barlach1990, abstract = {In this paper we give a classifier design approach, which yields classifiers of high computational efficiency and low memory requirements. The method is based on subdivision of the pattern space by a minimum number of optimized linear inequalities, i.e. the discriminators, into regions containing the pattern classes. The discrimination vectors are selected from the principal axes of each pattern class covariance matrix. We present an optimization procedure for sets of discrete parameter bounds. We briefly discuss the problem of multimodal pattern classes, and the possibility of using suboptimum solutions. The main objective is to design a group of optimum classifiers for payphone coin classification. In two examples we compare our classifier performance with the classification performance derived from fitting normal probability densities to the pattern classes.}, added-at = {2011-03-27T19:47:06.000+0200}, address = {New York, NY, USA}, annotation = {propably operates from sound because sciencedirect lists sound related publications as similar}, author = {Barlach, Flemming}, biburl = {https://www.bibsonomy.org/bibtex/21cd6d90b4b8d0707e782a429cef640c4/cocus}, doi = {http://dx.doi.org/10.1016/0031-3203(90)90024-F doi:10.1016/j.physletb.2003.10.071}, interhash = {3e2335e464188b8dec11f29bf5d006ec}, intrahash = {1cd6d90b4b8d0707e782a429cef640c4}, issn = {0031-3203}, journal = {Pattern Recognition ({PR})}, keywords = {Coin Decoupled Optimized classification; coins, covariance discriminators linear matrix;}, number = {3--4}, owner = {CK}, pages = {379--384}, publisher = {Elsevier Science Inc.}, timestamp = {2011-03-27T19:47:06.000+0200}, title = {Payphone coin validation using pattern recognition}, url = {http://www.sciencedirect.com/science/article/B6V14-48MPKH1-T1/2/f60fddbb8b5ad144210229e877f1ceac}, urldate = {2008.01.30}, volume = 23, year = 1990 }