Part-based handwritten numeral recognition

Seiichi Uchida, Marcus Liwicki, Vincent Marsault

In: Technical Committee on Pattern Recognition and Media Understanding. Pattern Recognition and Media Understanding (PRMU-2010) February 18-19 Tokyo Japan Seiten 1-6 2/2010.


In this publication we apply SURF for recognition of handwritten digits. In contrast to the general thought that global information is important for the recognition, we show that we achieve a recognition rate of more than 95 % by applying SURF in combination with KNN classifiers.


Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence