Spectral Modelling for multi-channel printers

Radovan SlavujPartner: HIG, Norway

Fellow researcher: Radovan Slavuj


Supervisor: Prof. Jon Yngve Hardeberg

Work Summary

Printing has been one of the leading information communication media for many centuries. Since the ink-jet photo printers are starting to be commercially popular, the quality of the printing has been raised to the highest standards. The high end ink-jet photo-printers are called multichannel or multi-ink printers as they usually have more than four process inks and some number of specialized inks. Colours that are possible to reproduce with these printers are very chromatic, vivid and durable.

Today we are used to computers and we send the image or a page to print electronically. Every time when one clicks the “print” button, the image passes though the processing before it is being printed. What we see on the display is not the same as what has been printed but the processing of the image aims to make it as close as possible. This processing is called colour management as it deals with numbers representing colours. Normally, this processing is per pixel, so each pixel (usually our camera or display has RGB pixel) is transformed to printer pixel (usually CMYK). Therefore, three numbers are sufficient to represent one pixel in the image. If one wants to print a physical property of the object (e.g. our sweater or jeans), the measurements of such material would contain around 35 numbers for each pixel. This is the real, physical measure of any real world object, also known as spectral reflectance. The image that contains spectral reflectance per pixel would be extremely difficult to process. Normal camera image takes around 5MB on hard drive, but the same image containing spectral reflectance per pixel would be more than 1GB, so the growth in complexity is exponential. For processing such image, usually printer models are used which process this image so that it can be printed on the CMYK printer. This is from e.g. 35 to 4 channel reduction and mathematically this represents an underdetermined problem. If the printer has more inks (channels), like the high end ink-jet printers have these days, the problem of channel reduction is solved more accurate, but it takes more time to process.

Radovan’s task in this project was to work on the improvements of spectral printer models for multichannel printers and possibly proposal of a new model. The initial strategy was to work with four channel printer models such as Neugebauer model and aim to extend the idea into multichannel printing. By using this idea and by targeting calibration and characterization of a printer, the improvements are suggested in characterization target estimation, calibration of the channels and the model. Three papers were made around this idea and one paper has been extended to a proposed new model.

The second topic in the work done by Radovan was concerned on modeling (characterization) and halftoning. He worked on establishing a link between these radically different areas of printing. Work in this area has resulted in newly proposed halftoning for multichannel printing and in the work that merges spectral reproduction workflow with multichannel halftoning.

In strive to achieve research goals, Radovan collaborated with other project members, particular Ludovic Coppel, Melissa Olen and Paula Žitinski Elías. The secondments were targeted to Fraunhofer FOKUS in Berlin, where useful inputs were gained from Dr. Jeremie Gerhardt who has also worked in similar

Publication and Dissemination

  • Slavuj, R., Marijanovic, K. and Hardeberg, J. Y. (2013), Feasibility study for textile colour simulation with multichannel printing technology, in `AIC Colour 2013′, International Colour Association (AIC), NewCastle upon Tyne, UK, pp. 379 382.
  • Slavuj, R., Nussbaum, P. and Hardeberg, J. Y. (2013), Review and analysis of spectral characterization models and halftoning for multi-channel printing, `Advances in Printing and Media Technology, Digitalization of Print’, IARIGAI print and media research, Chemnitz, Germany.
  • Slavuj, R., Coppel, L. G., Olen, M. and Hardeberg, J. Y. (2014), Estimating Neugebauer primaries for multichannel spectral printing modelling, SPIE Electronic Imaging, San Francisco, CA, USA.
  • Slavuj, R., Marijanovic, K. and Hardeberg, J. Y. (2014), Colour and spectral simulation of textile samples onto paper; a feasibility study, Journal of the International Colour Association, Vol. 12, pp. 36 − 43.
  • Coppel, L.G., Le Moan, S., Zˇitinski, P. E., Slavuj, R., Hardeberg, J. Y. (2014), Next generation printing – Towards spectral proofing, ‘Advances in Printing and Media Tech- nology, Print and Media Research for the Benefit of Industry and Society’, IARIGAI print and media research, Swansea, UK.
  • Slavuj, R., Pedersen, M., Hardeberg, J. Y. (2015), Multichannel DBS halftoning, SPIE Electronic Imaging, San Francisco, CA, USA.
  • Slavuj, R., Coppel, L. G., Hardeberg, J. Y. (2015), Effect of ink spreading and ink amount on the accuracy of the Yule-Nielsen modified spectral Neugebauer model, SPIE Electronic Imaging, San Francisco, CA, USA.
  • Workshop, Colour Printing 7.0: Next Generation Multi-Channel Printing, IS&T Color and Imaging Conference, Boston, Massachusetts, USA.

Leave a Reply