Copybook style determination of pseudo-online handwriting data
Abstract
Questioned document examination plays an important investigative and forensic role in many types of crime. For a ransom note or forgery, writer identification is the main goal. As an intermediate step towards writer identification, this dissertation presents a procedure for document examiners to determine the copybook style of a questioned document. Determining the copybook style helps to narrow the suspect population of the writer. The procedure uses a combination of pseudo-online data, the SDSS (Stroke Direction Sequence String) features, and the modified Levenshtein edit distance to obtain similarity matching scores. A second part of this dissertation focuses on cluster analysis of the pseudo-online copybook style database to determine what styles are similar. The resulting clusters group the similar copybook styles and indicate some geographic groupings.
Subject Area
Computer science
Recommended Citation
Manfredi, Mary L, "Copybook style determination of pseudo-online handwriting data" (2005). ETD Collection for Pace University. AAI3191872.
https://digitalcommons.pace.edu/dissertations/AAI3191872
Remote User: Click Here to Login (must have Pace University remote login ID and password. Once logged in, click on the View More link above)