Predicting the colorimetry of spot colour overprints
1. ICC/HP Digital Print Day, Sant Cugat, Spain
Predicting the colorimetry of
spot colour overprints
Kiran Deshpande, Phil Green
London College of Communication London
5. Objectives
• Predict the colorimetry of spot colour overprints
• Simple numerical method - easy to implement within ICC or PDF/X
workflow
• Free of intellectual property
• Application - previewing overprints of spot colours on-the-fly using
pre-media software
6. Proposed overprint model
• Reflectance of overprint ! product of the reflectances of two inks
measured independently
• Error is typically a linear underestimate of the reflectance
7. Proposed overprint model
• Reflectance product is modified numerically by a scaling factor
• Scaling factor depends on colorant opacity, ink sequence and dot area
9. Proposed overprint model
• Overprint colour correlated to the product of background and
foreground colours using regression analysis
• Refinement of linear model (Deshpande, 2010) to improve accuracy
X, Y, Z : predicted tristimulus values of the overprint colour
Xb, Yb, Zb : measured tristimulus values of the background colour
Xf, Yf, Zf : measured tristimulus values of the foreground colour
jx, jy, jz : scaling factors of the foreground colour depending on dot area
kx, ky, kz : exponents of the foreground colour depending on dot area
10. Evaluation method
• Design of the test chart (CMYK + Spot1 + Spot2)
• Total 4550 colour patches
0 25 50 75 100
0
25
+ + 50
75
=
100
CMYK (CMYK + 2 Spots)
Pantone157C Pantone330C
182 patches Block of 25 patches
(Spot1) (Spot2)
IT8.7/3 Basic set
11. Evaluation method
• Specifications
Printing Process Offset (Mitsubishi DAIYA 306)
1. Mitsubishi MYU Coat NEOS
Substrates 2. Hokuetsu Pearl Coat N
3. Nihon Be-7
Ink Sequence K - C - M - Y - PMS157C (Spot1) - PMS330C (Spot2)
Measurements X-Rite SpectroScan (ISO 13655:2009)
12. Evaluation method
• Calibrate the model-coefficients using the ink step-wedge chart
• Ink ramps of 0% to 100% over three different backgrounds
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%
White background
Grey background
Black background
Ink step-wedge chart - Pantone 330C (Spot2)
13. Results
• Data fitting - power regression
Measured overprint colour vs. product of the background and foreground colours
14. Results
• Overprint model - power regression
3
Average CIEDE2000
2.29 2.2
2 2.12 2.06 2.15 2.1
1.9 1.82 CMYK + SPOT1
1.76
CMYK + SPOT2
1
SPOT1 + SPOT2
0
Be-7 MYU Coat NEOS Pearl Coat N
Substrates
20. Implementation
• Multiple-inks combination: 3-inks, 4-inks and more
• Apply the model recursively
Spot3
Spot2
= = +
Spot1
3-inks 2-inks 3-inks
overprint combination combination
21. Implementation
• Process inks (CMYK) + Spot colour combination
Foreground Colour
Ink step-wedge chart (Xf, Yf, Zf)
of Spot Colour 2 & Coefficients (j and k)
Spot Colour 2
(Foreground)
Overprint Colour
(X, Y, Z)
Overprint
Model
CMYK
(Background)
CMYK + Spot Colour 2
Substrate
ICC profile Background Colour
'A2B' tag (Xb, Yb, Zb)
22. Summary
• Proposed model - predicts solid and halftone overprints
• Simple and computationally inexpensive - based on CIEXYZ
• Accuracy is acceptable for the overprint preview
• Implementation for different scenarios
• Applications
• previewing overprints in pre-media software
• matching spot colour overprints on digital printing systems
• implement within ICC or PDF/X workflow
23. Acknowledgement
• Thanks to Toppan Printing, UK for printing the test charts and
providing substrates for this project
24. References
• Chung, R., Riordan, M. and Prakhya S. (2008) Predictability of spot colour overprints,
Advances in Printing and Media Technology, Vl. XXXV, p. 373-380
• Deshpande, K. and Green, P. (2010) A simplified method of predicting the colorimetry
of sot colour overprints, Proc. 18th Color Imaging Conference: Color Science and
Engineering Systems, Technologies, and Applications, p 213-216, San Antonio, Texas
• ISO 13655(2009) Graphic technology - Spectral measurement and colorimetric
computation for graphic arts images
• Viggiano, J.A.S. and Prakhya, S. (2008) Prediction of overprint spectra using trapping
models: A feasibility study, TAGA 2008, Rochester, NY:2008 TAGA student chapter.