Printer forensics based on character image quality measures

Linjie Shen,Xiangwei Kong,Xingang You
DOI: https://doi.org/10.3321/j.issn:1001-0505.2007.z1.020
2007-01-01
Abstract:A new printer forensics method based on character image quality measures is put forward to identify the source of printed files.Based on the analysis of the properties of the "banding" noise in the character image extracted by Gaussian filter,the statistic features donating the characteristics of each printer are extracted using image quality measures(IQM),and support vector machine(SVM) is used as classifier to identify the source printers.Compared with other methods,the printers used in the experiment are all identified correctly and the accuracy of character identification is much higher with the proposed approach.The experimental results prove that the individual characteristics of each printer can be identified effectively by the character image quality measures.
What problem does this paper attempt to address?