Optimum Design Parameters of the Classifiers for Omni-Font Machine-Printed Numeral Recognition Based on the Minimum Classification Error Criterion
Katsumi Marukawa, Kazuki Nakashima, Hiroshi Shinjo, Yoshihiro Shima, Hiromichi Fujisawa · Machine Vision Applications · 1998
The optimal design parameters of classifiers for omni-font machine-printed numeral recognition based on the minimum classification error (MCE) criterion are determined experimentall y. The design parameters that influence the accuracy of an optical character reader (OCR) are: similarity measure (or distance measure), kinds of features, dimension of the feature vector, method of training, number of templates percategory, and the size of a training sample set. It way found that the optimum &sign parameters were simple similarity, four templates per category, and 576 dimensions (i.e., four directional feature planes of 12 x 12 blocks). The directional feature classifier with these design parameters gave the best performance and had the smallest memory size and computational cost of all the classifiers.