Abstract not found

We evaluate different two-dimensional non-linear deformation models for handwritten character recognition. Starting from a true two-dimensional model, we derive pseudo-two-dimensional and zero-order deformation models. Experiments show that it is most important to include suitable representations of the local image context of each pixel to increase performance. With these methods, we achieve very competitive results across five different tasks, in particular 0.5 % error rate on the MNIST task. 1

For handwritten character recognition, a new elastic image matching (EM) technique based on a class-dependent deformation model is proposed. In the deformation model, any deformation of a class is described by a linear combination of eigen-deformations, which are intrinsic deformation directions of the class. The eigen-deformations can be estimated statistically from the actual deformations of handwritten characters. Experimental results show that the proposed technique can attain higher recognition rates than conventional EM techniques based on class-independent deformation models. The results also show the superiority of the proposed technique over those conventional EM techniques in computational efficiency.

In online character recognition based on elastic matching, such as dynamic programming matching, many of misrecognitions are often caused by overfitting, which is the phenomenon that the distance between reference pattern of an incorrect category and an input pattern is underestimated by unnatural matching. In this paper, a new recognition technique is proposed where category-specific deformations, called eigen-deformations, are utilized to suppress those misrecognitions. Generally, matching results at overfitting are not consistent with the eigen-deformations. Thus, the overfitting can be detected and penalized by a posterior evaluation of this inconsistency. The result of a recognition experiment showed the usefulness of the proposed technique. 1

Hiermit versichere ich, dass ich die vorliegende Diplomarbeit selbstständig verfasst und keine anderen als die angegebenen Hilfsmittel verwendet habe. Alle Textauszüge und Grafiken, die sinngemäss oder wörtlich aus veröffentlichten Schriften entnommen wurden, sind durch Referenzen gekennzeichnet. Aachen, im Juli 2008

Abstract not found