|
Ordered vector set for binary image description and an application to handwritten character recognition set AbstractIn this paper a new algorithm to describe a binary image as a ordered vector set is presented. We define an extension of a string edit distance to compute this ordered set edit distance used in nearest neighbor classification task. Finally, the advantages of this method applied to isolated handwritten characters classification are shown, compared to similar methods based on string or tree representations of the binary image.
[Edit] |