Experiments with new raster elements.

Ziljak Vujic, Jana ; Plehati, Silvio ; Kropar Vancina, Vesna 等

Abstract: In this work new unpublished raster elements (RE) are tested in security and individualized printing. The results of earlier researches dealing with graphic technology and design are published at the conferences. Results obtained by measuring outline of RE, with defined coverness area and coverness area after digital printing are given in this work. For new RE, as well as for earlier published RE no discussions and/or researches of these parameters are still not done. Measuring procedure and valorization of regular behavior or change of RE shape in the whole area of coverness are suggested. This procedure should be used whenever new RE is introduced in praxis.

Keywords: new raster element (RE), new procedure of measuring and validation of RE

1. RASTER ELEMENT "MUTANT M68"

Complex mutant in this work is given according to relation as follows:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

ContourPlot[1-z,{x,-1,1},{y,-1,1},Contours[right arrow]8]; z=Abs[ Sin [ Abs[(0.25^(x^2)-y^2) ]] ]/1.5

[FIGURE 1 OMITTED]

It is difficult to prepare PostScript code from the figure of two-dimensional shape. It is necessary to add relation with angle arrangement and divider of the whole equation. It is even more difficult to find out relation with three-dimensional shape. The only way to find out errors in definition of RE is to use 2D and 3D analysis. It is not possible to find dislocation of definition from space -1,1 in all three coordinates by RIPing but it can be done by 2D and 3D views. Such views ensure reproduction realization by digital printing or CTP processes. Even checked PostScript units do not give evidences about errors during process of screening (Ziljak & Pap, 1999).

Authors suggest to keep uncompleted values of coordinate z while trying to find new RE. Even the smallest penetration of this coordinate over values -1 and 1 can cause stopping/disconnection in digital screening. When new REs are requested it is suggested to provide parallel tests of relations in 2D, 3D and PostScript technique.

Plot3D [z,{x,-1,1},{y,-1,1},PlotRange[right arrow]{0,1}]

[FIGURE 2 OMITTED]

Research of new RE include experiments with coverness of color reproduction with RE in continuous gray levels in low and higher frequencies (Ziljak et al, 2003). In this work the quality determination method is established and it has to be used practically always during creation of new RE shapes which have never been used in printing praxis before. Researches are related to regular coverness, changes of raster angles and results obtained at various frequencies. The same research is enlarged to cover digital printing because these techniques are expanding now (Ziljak Vujiae et al., 2006).

In the figure 3 structure of new RE M68 with 50% coverness after RIPing with sharp edges is given. Printing plate and process of printing have added some enlargement of RE at the edges which is for the same example presented in figure 4. For realization PostScript graphic language was used (Ziljak & Pap, 1999; 2003). Converting of earlier relation gives following shape:

/r68 {dup mul exch 2 exp 0.25 exch exp exch sub 2 div abs 90 mul sin abs 3 div } bind def

[FIGURE 3 OMITTED]

2. PRINT WITH MUTANT RASTER ELEMENT

Prints with s 900 dpi, 1200 dpi and 1800 dpi were scaned with the same number which was used for bitmap screening of figures to determine coverness of prints. By this technique it is possible to research dot gain in RE neighbourdhood, because it is possible to read brightness values for each scaned pixel. If all brightness values of scaned pixels in researched area are added together and obtained value is divided by the number of researched pixels mean value of brightness for researched area is obtained. Digital print of Mutant RE - r68 (M68) with 20 dpi printed by "Xeikon 32" is presented in figure 4 in two enlargements. Upper part of print covers wider space and bottom parth represents space of one raster cell. Table 1. presents parts of brightness (and darkness) for individual pixels after scanning.

[FIGURE 4 OMITTED]

Values of density for some edge-parts of RE in table 1 are given in 10 steps and separated for coverness of 100% and coverness of 0%. There are differences in overall coverness between faze of RIPing before printing plate is made and after printing. Results of these measurements are given for M68 in Table 1. The first column in table 1 represents marks for L 20 lpi where this test was done and defined covernesses of 10, 30, 50, 70 i 90% (abbr Xla20 10). The second column represents number of scanned elements of prints which have entered calculation of density and coverness.

A lot of pixels (more than half of million) were included in research of space and resolution of scanning. Intern structure of every raster element in raster cell was researched when every raster element was divided in hundreds of parts. In the frame of dot gain research it was possible to analyze all gray areas around RE for new mutant shapes, especially for complex crystal and lace raster.

White area was changed with gray area. At high coverness white area disappears and black area is drastically smaller. Color/ink dispersion from the center of defined agglomeration is bigger for the bigger defined coverness. Gray areas gradually and completely close white space.

[FIGURE 5 OMITTED]

In table 2 coverness of mutant M68 before printing is given (RIP B/W), as well as after digital printing where in calculation are included all parts of gray areas in the neighborhood of RE and in the space of raster cell.

For mutant M68 deviations at 20 lpi are very small. The figure is defined to the minimal zero deviation after print scanning.

3. CONCLUSION

Introducing of new forms of screening will be used in security printing. Suggested mutant with changeable structure is not safe enough because it has two sub-shapes smaller and bigger central structure. It can be used for documents and design of posters and ambalage. In the frame of this work 2D and 3D shapes are given, behavior of raster in continuous spreading, as well as in areas with disconnections of pixels in various degrees of density. The behavior is the same as it is when conventional raster shapes are used.

4. LITERATURE

Ziljak, V., Pap K. (1999). Mathematical model of stochastic curves in digital printing, 26th International Research Conference IARIGAI, Munchen

Ziljak-Vujiae J., Pap, K., Ziljak, I. (2006). Modeling of screening elements in stochastic multi-color reproduction, 33rd International Research Conference IARIGAI, Leipzig

Ziljak, J., Vaneina, V., Agiae, D., Ziljak, I., Pap, K. (2003). New screening elements in multi-colour printing for special purposes, 30th International Research Conference IARIGAI, Dubrovnik - Croatia

Ziljak, V., Pap, K., PostScript, (1999, 2003). FS, ISBN 953-199-00-X, UDK: 681.3.06;655.26, Zagreb

ZILJAK VUJIC, J[ana]; PLEHATI, S[ilvio] * & KROPAR VANCINA, V[esna] *

Table 1. Relative participation in darkness--percents
of total number of pixels in percents of darkness

Data Pixels Black % White% 1-10 10-20

XLa20 10 518400 3.080 83.563 1,8245 1,382
XLa20 30 518400 22.366 56.510 2,1956 1,663
XLa20 50 518400 33.508 27.533 2,9971 2,536
XLa20 70 518400 47.459 0.345 5,1105 4,012
XLa20 90 518400 73.483 0 5,6105 4,792

Data 20-30 30-40 40-50 50-60 60-70

XLa20 10 1,084 0,9898 0,899 0,8729 1,026
XLa20 30 1,452 1,4583 1,466 1,5945 1,881
XLa20 50 2,360 2,5068 2,451 2,6298 2,999
XLa20 70 3,611 3,7292 3,498 3,7546 4,523
XLa20 90 4,496 4,5934 3,680 2,4115 0,813

Data 70-80 80-90 90-99

XLa20 10 1,2986 1,565 2,407
XLa20 30 2,4481 3,248 5,715
XLa20 50 3,9921 5,498 10,987
XLa20 70 6,3499 9,706 7,906
XLa20 90 0,1163 0,001 0,000

Table 2. Coverness of RIPing and coverness of prints

data %
defined coverness
coverness of print gray edges deviation

L20 10 10 97,107 -0.2893
L20 30 30 296,882 -0.3118
L20 50 50 479,417 -2.3118
L20 70 70 696,083 -0.3917
L20 90 90 927,116 2.7116

data
defined %
coverness coverness RIPing

L20 10 103,840 0.3840
L20 30 300,290 0.0290
L20 50 499,269 -0.0731
L20 70 699,331 -0.0669
L20 90 901,436 0.1436