Title :
Learning Dirichlet kernel histogram functions for pattern recognition
Author_Institution :
Kyushu Univ., Fukuoka, Japan
Abstract :
A formula of approximating histograms is presented. The formula is an explicit function of data permitting the inclusion of unknown parameters. The parameters in the formula are learned so as to classify training data to build a pattern classifier. Recognition is performed by applying the classifier to testing data. Our method is used for the recognition of vehicle-type.
Keywords :
image classification; learning (artificial intelligence); object recognition; Dirichlet kernel histogram functions; pattern classifier; pattern recognition; vehicle-type; Abstracts; Histograms; Testing;
Conference_Titel :
Signal Processing Conference, 2009 17th European
Conference_Location :
Glasgow
Print_ISBN :
978-161-7388-76-7
