Research of mutant screen element stochastic application.

Ziljak, Ivana ; Mrcelic, Zeljka ; Dujic, Lucijana 等

1. INTRODUCTION

This article proposes a new manner of creating areas of programmed screen shapes (Foley at al., 1997) in order to select shapes from this area by the stochastic selection, as well as to define angles from 0 to 90 degrees and screen rulings in the acceptable range of screen rulings (Adobe Systems I., 1999) of a printer. The creation of PostScript algorithms that, depending on the seed generator of random numbers implement pseudorandom N of individual screen shapes, angles and screen rulings, has resulted in a unique print by using one single number (SEED). Stochastic application of newly developed mutant screen elements (Pap at al., 2008) demands preliminary testing.

2. MUTANT SCREEN ELEMENT R73 IN GRAY LEVEL AND ANGLE TESTING

The software solution has been made with functions that include the transformation model that can be generated either continuously or stochastically. The screen R73 has been tested in detail in the definition area (Fig. 1). Low screen ruling are analyzed by print scanning. Some print scanning has been done with 900 dpi to study the dot gain in the environment of individual screen element (Fig. 2).

Experiment results of testing mutant screen R73 for different gray levels and different angles are presented in figures 3, 4, 5 and 6. A rough screen ruling of 3 lpi is used in all the examples in order to observe the parameter mutation development in the best possible way. Each row represents one mutant.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

The zero mutant is in the first row where the mutant parameter amounted to 0, and in the tenth row the mutant parameter is 1. On basis of analyzing such experimental results under low screen ruling and various angles, decision is made as to the allowed initial and final mutation shapes, and even as to discrete mutations, if necessary.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

3. STOCHASTIC APPLICATION

The stochastic part of the discussion refers to application of an amplitude coverage model of each pixel separately (Ziljak & Pap, 2000). This applies to color printing as well where the usual procedure is to set various angles of screen. As in this case, a randomly chosen angle is joined with each pixel, there is no limiting in the number of colors: it is possible to mix process and spot inks.

Algorithms enable connecting of the color image with the angle and ruling screening alterations. We extend our research work with new screen element R73. The goal is to achieve full individualization of reproduction with the intention of applying in security printing. There is dot gain of each element's environment after printing, so the real border could not be determined any longer. Care should be taken only in respect to achieving the same coverage level: before printing and after printing.

Stochastic application of R73 halftone screen in 3 prepared images is presented in figures 7, 8 and 9. In this manner three stochastic parameters have been used in the Postscript program: /kory 0.7 rn mul 0.2 add def % mutat. param. from 0.2 to 0.9, /L {15 15 rn mul add} def % screen ruling from 15 to 30, /K {0 90 rn mul add} def % screen angle from 0 to 90.

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

A novelty in this paper is the processing methodology in terms of the PostScript RIP printer. It implies that data on the input image move with the algorithm to RIP (Brett, 2000) and the SEED parameter that individualizes the image, generates a chain of random numbers from the RIP random number generator and not from the input PC. This is how the print becomes individualized not only by algorithms of new screen shapes, but also by a RIP random number. It is shown in our examples that the some SEED number is used.

4. CONCLUSION

A stohastic solution of print individualization is a border area of digital printing. Changes in PostScript commands enable a programming intervention during printing. Firstly, each printing sheet can have a new number, different image or text, prepared in a separate file. Secondly, individualization may be generated by an algorithm (commonly used in numeration), meaning that printed data are not in the memory, but created by logic defined in the PostScript program. This article proposes a method of randomly selected parameter mutant shape of a screen element for each pixel and presents the highest degree of individualization.

If various SEED values are defined for each input color chanel, it is possible to achieve multiple combinations of protection against counterfeiters. Program solution is created in a way which can make each researcher interested in it, continue the development of his own algorithms, procedures and finally apply it in specialised usages.

5. REFERENCES

Adobe Systems I., (1999). PostScript Language Reference, third edition, Adobe Systems Incorporated, pp. 482, ISBN 0-201-37922-8

Brett, G. (2000), Digital Prepress Technologies, Pira International Ltd, pp. 122, ISBN 1-85802-261-4

Foley, K.; van Dam; Feiner & Hughes (1997). Computer Graphics Principles and Practice, pp. 569, Addison Wesley, ISBN 0-321-21056-5

Pap K.; Ziljak I. & Ziljak Vujic J. (2008). Design of Digital Screening, pp. 73, FotoSoft, ISBN 978-953-7064-10-5, Zagreb

Ziljak, V. & Pap, K. (2000). Mathematical Model of a Stochastic algorithm for Digital Printing, Advances in Printing Science and Technology, pp. 295-302, ISBN 1-85802-355-6, Leatherhead, 2000, Pira International Ltd

ZILJAK, Ivana; MRCELIC, Zeljka; DUJIC, Lucijana; MARINCEL, Petra & ZILJAK VUJIC, Jana *

* Supervisor, Mentor