Printer Identification Based on Distance Transform

Printed document examination plays an important role in public security. A printer identification system is presented for the purpose of identifying the printer which created a suspect printed document. It is composed of image acquisition, image preprocessing and character matching. An optical instrument is developed for document image observation and acquisition under high resolution. Normalized printed characters are obtained and recognized in preprocessing. Distance transform algorithm is applied to character image matching and distance between printed documents is calculated. Finally the minimum distance classifier is applied to printer identification. Experiments are carried out in a database including 45 laser printers and very promising results are achieved.