Fingerprinting Laser Printers

It’s not just about little yellow dots. The variations in the digital-to-paper chain inherent to a laser printer are different enough from model to model that one printer’s banding is unique enough to identify it.

Errors like drum RPM, gear eccentricity and backlash, and polygon mirror wobble create a unique banding ‘signature’. This is readily apparent when you print large blocks of black: there are very few laser printers out there that can reproduce a big black square as a perfectly flat black area. However, this banding is just as present in text… you just don’t notice it.

Researchers have figured out how to reconstruct banding by isolating the text from a high-resolution scan, and from there generate the printers’ signature.

https://engineering.purdue.edu/~prints/public/papers/icme07-nitin.pdf

“Forensic characterization of devices is important in many
situations such as establishing the trust and verifying authen- ticity of data and the device that created it. Current foren- sic identification techniques for digital cameras, scanners and printers are highly reliable due to the fact that each of these devices cannot escape inherent electro-mechanical properties which add “signatures” to the data they produce.
[…]

Today there are
two primary technologies for desktop printers – electrophoto- graphic (usually laser) and inkjet. The very same features that give rise to an intrinsic signature for these devices may also cause visible and unacceptable image artifacts if they are not properly controlled.

Figure 2 shows a side view of the cartridge for a typical
EP printer. The print process has six steps. The first step is to uniformly charge the optical photoconductor (OPC) drum.
Next a laser scans the drum and discharges specific locations on the drum (exposure). The discharged locations on the drum attract toner particles (development) which are then attracted to the paper which has an opposite charge (transfer). Next the paper with the toner particles on it passes through a fuser and pressure roller which melts and permanently affixes the toner to the paper. Finally a blade or brush cleans any excess toner from the OPC drum.

In EP printing, artifacts are created in the printed output
due to electromechanical imperfections in the printer such as fluctuations in the angular velocity of the OPC drum, gear
eccentricity, gear backlash, and polygon mirror wobble. In
previous work we have shown that these imperfections are di- rectly related to the electromechanical properties of the printer. This property allows the corresponding fluctuations in the de- veloped toner on the printed page to be treated as an intrinsic signature of the printer. The most visible print quality defect in the EP process is banding, which appears as cyclic light
and dark bands.

Techniques that use banding in electrophotographic (EP)
printers as an intrinsic signature to identify the model and manufacturer of the printer have been reported in [6]. It is shown that different printers have different sets of banding frequencies which are dependent upon brand and model. This
feature is relatively easy to estimate from documents with
large midtone regions. However, it is difficult to estimate the banding frequencies from text. The reason for this is that the banding is present in only the process direction and in printed areas. The text acts as a high energy noise source upon which the low energy banding signal is added.

One solution which was previously reported in [7] is to
find a feature or set of features which can be measured over smaller regions of the document such as individual text char- acters. If the print quality defects are modeled as a texture in the printed areas of the document, then texture features can be used to classify the document. These types of features
can be more easily measured over small areas such as inside
a text character. The approach in [7] was based on texture
measures estimated from the graylevel co-occurrence matrix
(GLCM) generated from the printed region of the text charac- ters. A support vector machine (SVM) is then used to select
the printer most likely to have printed the page. Each text
character that is processed in the document casts a vote for the most likely printer and the majority vote is taken as the final decision. This framework provides very robust perfor-
mance.”

Advertisements
%d bloggers like this: