Prediction of laser printers and cartridges based on three-dimensional profiles via discrimination analysis
Zi-Feng Jiang,Qing-Hua Zhang,Ya-Chen Wang,Yan-Ling Liu,Ya-Wen Zhao,Yu-Yu Hao,Jing-Yuan Xu,Xu Yang,Xiao-Hong Chen
DOI: https://doi.org/10.1016/j.forsciint.2024.112186
Abstract:Printer source prediction is an important task when examining questioned documents. While some research has provided methods to predict the source printer of documents, with the advent of compatible consumables, printer prediction could become more complex and difficult. Predicting the source printer after replacing cartridges and identifying the source of printer cartridges are unresolved issues that are rarely addressed in current research. Herein, we introduce a novel technique to predict the manufacturer, model, and cartridges of laser printers (i.e., compatible, and original cartridges) used to produce a given document. Document samples produced using eight laser printers and 247 cartridges were collected to establish a dataset. Common manufacturers included HP, Canon, Lenovo, and Epson. After obtaining white-light images and three-dimensional profile images of printed characters, a morphological analysis was conducted by questioned document examiners (QDEs) using microscopy. Microscopic image features across a series of images were also extracted and analyzed using algorithms. Then, six high-dimensional reduction algorithms were used to obtain between- and within-printer variations as well as between- and within-cartridge variations. Finally, we conducted principal component analysis (PCA) and discriminant analysis. For 40 % of the samples, mixed discrimination analysis (MDA) and fixed discrimination analysis (FDA) were employed to predict the manufacturer, model and cartridge of laser printers used to produce the questioned printed document; the remaining 60 % samples comprised the training dataset. In the prediction of manufacturer, model and cartridge, our method achieved mean accuracies of 95.5 %, 97.5 %, and 90.2 %, respectively. Hence, this technique could reasonably aid in predicting the manufacturer, model, and cartridge of a laser printer, even if different cartridges are loaded into printers.