Joel Zlotnick, Jordan Brough and Troy Eberhardt
Line art, microprinting, guilloche patterns, security halftones, transparent register and other features based on printed artwork are common counterfeit deterrence features based on design, instead of technology. However, such printed features either require the technical capabilities of a security printing press to execute, or resist redrawing due to high image complexity, with limited emphasis on defeating counterfeiting in prepress. The authors contemplate a different function for security document artwork, which can be designed to interrupt certain prepress steps inherent in traditional counterfeiting workflows by blocking a would-be counterfeiter from understanding the press set-up used to print a genuine document.
Counterfeiting of security document artwork follows one (or both) of the simplified graphic arts workflows summarised in table 1. The scan-and-print process of digital counterfeiting is popular because of its simplicity, though technically limited because inkjet and toner devices can only simulate line artwork with process colours and halftones. In contrast, a traditional counterfeiter attempts to replicate details in genuine document artwork, including painstakingly redrawing the design on each of the different printing plates and printing the counterfeit using a press set-up such as the one used to print the target genuine document. The traditional counterfeiting workflow has the potential to produce a better quality fake, even though it requires more time, more sophisticated graphic arts skills, and more expensive printing equipment.
Comparison of steps in digital and traditional counterfeiting workflows:
|Simplified digital counterfeiting workflow||Simplified traditional counterfeiting workflow|
|Obtain a template genuine document||Obtain a template genuine document|
|Scan or photograph the genuine document.||Separate/isolate printing plate images from one another.|
|Edit digital image in raster image editing software.||Rebuild plate artwork in vector image drawing software.|
|Output counterfeits using CMYK office equipment.||Manufacture offset and/or intaglio printing plates.|
|Output counterfeits on press, in line art and
Consider the second and third steps of the traditional counterfeiting workflow in table 1, in which the counterfeiter must separate the printing plate images within a genuine document as a precursor to redrawing the artwork. The separation of images may be accomplished with the help of digital imaging software, or with visual examination; either way, the counterfeiter’s goal is to understand how many printing plates were used, and the specific artwork and colours of ink used on each. Without this information, there is no way to accurately replicate the original artwork and colour arrangement for each print plate.
How do counterfeiters separate and count plate images? What microscopic cues in genuine documents facilitate the ‘decoding’ of document artwork? What artwork design techniques can be used to hinder the process? The concepts are best explained through examples. The reverse engineering examples described below are shown as enlargements of microscopic areas for simplicity, but obviously traditional counterfeiters must reverse engineer the entire document artwork. Horizontal colour bars have been added to the lower portion of most of the figures to assist the reader in identifying the spot colour selections and colour transitions in the artwork. These colours are approximations of the actual ink colours and include only background artwork. Since this discussion is limited in scope and is not intended to be a critique of any document, the authors intentionally omitted identification of the example documents to maintain a focus on the concepts.
In most printing workflows, ink of one colour originates from a single print plate. Figure 1 shows a green and light green geometric line pattern, which suggests this design was produced by two plates, one for each colour.
Continuity of lines in artwork is another important visual cue, and in Figure 1 the lines within each colour form a continuous, unbroken pattern. This background artwork is protected by an amplitude-modulated security half-tone that modifies the line widths. Importantly, this security halftone impacts the difficulty of redrawing the image, but not the difficulty of distinguishing between the plate images.This illustrates how impeding plate separation requires an entirely different design strategy than impeding artwork replication.
A single offset print plate usually delivers just one ink colour, except for the use of split fountains, which smoothly blend two or more colours of ink on press before inking the plate. The visual effect in the final print is a seamless colour transition across the artwork in a single direction (usually horizontal). The significance of this technique is that a multicoloured image was created from a single plate. If the split fountain artwork contains continuous lines or another repeating pattern, the colour transition can be visually traced as if the ink were a single colour. For example, Figure 2 was printed with two plates, but it contains more than two colours of ink.
One pattern is printed only with a light blue ink; the continuity of the lines and the consistency of the colour confirm this pattern is from a single plate. However, the second line pattern slowly changes from a darker blue at the left, to purple in the centre, and finally pink at the right. The continuous lines in the blue-pink line pattern reveal that it origi-nates from one plate, even though the art contains several distinct ink colours. The extra colours from the split add complexity, but are not designed to prevent reverse engineering of the plate artwork.
Some security documents add complexity by employing two, three or even four split fountain images. The example in Figure 3 shows many colours of ink, and contains three separate split fountains: purple to dark green, yellow to light green and blue to orange, from left to right.
This background design is a product of three separate printing plates and six inks. Each colour can be traced from left to right by following the unbroken lines in the artwork. Also consider how the artwork shapes are different for each plate and correspond to the ink colours. Despite the many colours present, this example contains many visual cues that help an observer understand how the design was created. Split fountain transitions are usually regarded as press capabilities that not all counterfeiters can mimic. How-ever, split fountains can also be used to print the same or similar colours of ink from two or more plates to confuse the plate set-up. For example, Figure 4 requires more than just an examination of colour to understand the plate set-up. The right side of the enlargement shows a dark orange ink in the design of ‘C’ shapes and a light orange ink in the connected line pattern which encompasses the ‘C’ shapes. However, both inks are part of two different split fountain transitions that are a similar grey colour on the left of Figure 4. As the left side of Figure 4 suggests only a single plate because there is only one colour, a viewer has to study the colours throughout Figure 4 to understand that the total artwork is from two different plates.
The use of the same colour of ink (grey) on more than one plate complicates a counterfeiter’s use of software colour selection tools to isolate the plate images, because in this example it is no longer true that each plate contains ink colours that are visually different from the other plates, as was the case in Figures 1 through 3. Consider the implications of making this approach the centrepiece of a security design strategy. On a press with several fountains and several splits per fountain, artwork could be designed to deliberately confuse the plate set-up by printing the same or similar colours of ink for multiple plates and using the split fountains not only to add colours, but instead to muddy the association between plate artwork and color. Other ways to hinder reverse engineering include decoupling the ink colours from the artwork. For example, instead of one plate containing the ‘C’ shapes and the second containing continuous thin lines, both shapes could be commingled on each plate with breaks in the continuous line pattern to impede line tracing.
The background print shown in Figure 4 contains two plate images that share a colour of ink, but the complementary elements from each plate are independent and do not overlap spatially. In Figure 5, the artwork contains two overlapping line patterns, one from purple to pink, and the other from brown to blue.
The colours in the centre of the background design are sufficiently similar to complicate separation based on colour alone. The easiest path to decoding this pattern lies at the left and right edges, where the colour differences help predict the line patterns used on each plate. The line patterns can then be traced through the rest of the design to understand the changing colours in both of the splits, and demystify the more confusing centre of the design. To make reverse engineering more difficult, the associations between colour and artwork could be delinked, and frequent breaks introduced along the horizontal pattern to prevent tracing.
For example, Figure 6 illustrates a design that integrates breaks to interrupt line tracing. First, each of the geometric shapes is separated from the others by non-printed areas of white paper, so would-be counterfeiters cannot simply follow lines to trace out a plate image. Secondly, this design employs a split fountain to transition from a bright, vibrant orange at the left to a darker orange at the right; this split between two closely related colours raises some very interesting counterfeit deterrence possibilities that will be explored more fully in Part 2 of this article series. Thirdly, the designer has manufactured the illusion of additional spot colours by registering similar shapes from two different plates. As seen in Figure 7, slight misregistrations of the shapes reveal that what appears to be a single dark orange (#1) spot colour ink is actually overlapping green and orange spot inks.
Similarly, spot blue and orange are overlapped to create a dark green (#2), and what looks like teal (#3) is actually overlapped blue and dark green shapes. These overlapping shapes, which contain hard edges and small registration variations, may provide clues to the counterfeiter that these new colours aren’t the result of a separate plate image.
Observing small misregistrations to identify the specific plate set-up is harder in the next example, which uses fine rough-edged parallel line patterns instead of geometric registered shapes as seen in Figure 7. This design, in Figure 8, contains numerous diamond shapes from different print plates that share a common colour gamut (such as in Figures 4 and 5), as well as art printed from different plates that overlap spatially (as in Figures 3, 5, 6 and 7). Decoding Figure 8 is difficult because the overall appearance of each of the diamond shapes is a product not just of multiple ink colours from different plates, but also of varying line thicknesses that allow ink coverage, and image density, to be varied within a single colour.
Additionally, the overlap of one ink with another creates the illusion of additional spot colours without the need for perfect registration. There are no large continuous patterns in this design and each diamond shape is bounded by a non-printed area of white paper, which prevents the tracing of lines. The patterns of parallel lines are oriented in various directions within each diamond shape; therefore, it is difficult for an observer to rely on the line direction or other facets of the artwork to understand the plate set-up. Split fountains are indeed used in this design, but are extremely difficult to identify at this scale and with this particular artwork. As in Figures 6 and 7, the macrovisual effect is, instead, of an increased number of discrete spot colours. Finally and importantly, the use of fine line patterns instead of block shapes makes the edges of the diamond shapes indistinct compared to the hard edges of the designs shown in Figures 6 and 7. This makes this design more tolerant of small variations in registration on press and prevents a would-be counterfeiter from using minor registration variation as a cue to understanding how the document was printed.
Overall, the artwork shown in Figure 8 prevents a more challenging reverse engineering problem than Figures 1 through 7 and introduces several valuable strategies, but even Figure 8 does not fully explore the potential of design strategies to inhibit artwork reverse engineering. The next question is; how to expand upon the examples discussed above to identify novel design strategies that specifically target prepress steps in traditional counterfeiting workflows. That topic will be explored in more detail in Part 2 of this article series.
When viewed through the eyes of a counterfeiter attempting to understand how a genuine document was printed, the examples described in this article demonstrate that design choices in security document artwork are not only cosmetic considerations. Choices about how colour, split fountains and artwork are implemented can have significant counterfeit deterrence implications in disguising the press set-up used to produce a genuine document. However, the examples presented here do not explore the full capabilities of designers to create security document artwork that exhibits the highest degree of resistance to reverse engineering. An expanded range of design techniques are possible that were not described here. Part 2 of this series will describe additional artwork strategies related to colour, ink density and split fountains that can further improve the ability of security designers to combat artwork replication by traditional counterfeiters.
MORE ABOUT THE AUTHORS
Joel Zlotnick is employed by the US Department of State, Bureau of Consular Affairs, Counterfeit Deterrence Laboratory as a physical scientist. He conducts research on how design strategies can help maximise the security value of printing technologies and security features, and develops training programmes on counterfeit detection.
Jordan Brough is employed by the Homeland Security Investigations Forensic Laboratory as a forensic document examiner, specialising in adversarial analysis and counterfeit deterrence. Jordan spends his time examining suspect documents and consulting with United States security document designers.
Troy Eberhardt is employed by the US Immigration & Customs Enforcement Homeland Security Investigations Forensic Laboratory. He supervises the Research and Development Section at the laboratory, which specialises in identifying and mitigating vulnerabilities within travel documents.