In today's world of highly sophisticated technological crimes, criminals including counterfeiters, document forgers, and other parties interested in altering information have a low barrier of entry. To combat these crimes requires developing a level of forensic analysis to aid law enforcement agencies in tracing the origins of documents or materials in question. In this article, the authors use printer forensics in an effort to understand the effect of resolution and character selection on the accuracy of printer identification. Specifically, they use a multiclass ADABOOST classifier to determine which of six printers, representing several ink jet and laserjet models, were used to produce a subsequently scanned image. Their results, investigating six different English characters, show that classification accuracy continues to increase with scanning resolution up to 1200 pixels/in. The results are character dependent, suggesting that different characters may be used for different forensic purposes–printer model, cartridge, and individual printer identification as examples.
Jason Aronoff, Steven Simske, "Effect of Scanner Resolution and Character Selection on Source Printer Identification" in Journal of Imaging Science and Technology, 2011, pp 50602-1 - 50602-10, https://doi.org/10.2352/J.ImagingSci.Technol.2011.55.5.050602