Automatic handwriting identification based on the external properties of the samples

A handwriting sample is identified by comparing its features to the corresponding features of a list of handwriting samples contained in a reference library. Some of the needed features are extracted from the structure of specific parts of the handwriting´s sample which have to be identified first (e.g., specific letters). Other features are extracted from the external structure of the handwriting (e.g., margins). Due to their described property, these features are extracted automatically with relative ease. Techniques for extracting features of the described kind are presented in this work. Experimental results obtained when using the extracted features for identification purposes are then given.