Summary

The physical model of colorimetric characterization on three-layers-stacked ChLCD is failed because its reflectance spectra have no scalability. Then, the numerical model with polynomials of three, four, six, eight, nine, eleven, fourteen, and twenty terms were used to characterize ChLCD. But the ΔE00_avg. between the

60

predicted values by polynomials and the measured values of 900 testing sets are all behind one. So, numerical model with regression is not suitable for characterizing ChLCD. Finally, look-up tables are utilized to do the colorimetric characterization on ChLCD. The ΔE00_avg. of various four geometrical interpolation methods in cubic of 9 non-uniform packing are all below 1. This result proved that look-up tables with interpolation are the adequate method to characterize ChLCD. Furthermore, among these four interpolation methods, the most beneficial method is tetrahedral interpolation.

61

Chapter 5

Conclusions and Discussions

5.1 Conclusions

As the green issue has been thought of an important theme for discussion, the need for e-Paper will grow with each passing day. Full color e-Papers are the main stream of research now. When different types of e-Papers have come to maturity, the color mechanism of them maybe would become more and more complex.

The objective of this thesis is to construct the colorimetric characterization of e-Paper, and ChLCD is taken for our example. At first, the reflectance spectra of ChLCD were proved that they do not have scalability because of its complicated color performing system, so the physical model of colorimetric characterization is failed to present the color properties of ChLCD. And then, the numerical model on ChLCD would be shown next. Finally, the best method which is look-up table with interpolation to characterize ChLCD would be taken on.

Eight different polynomials with three, four, six, eight, nine, eleven, fourteen, and twenty terms are utilized for the regression, and the unequally spaced lattice points (729 points) in the RGB space are taken to be the training sets in the numerical model. Firstly, the color differences between the predicted values of regression polynomials of 729 training sets and the measured values would be calculated.

Secondly, the color differences between the predicted by regression and the measured values of 900 randomized test samples in uniform color space would be calculated, too. Finally, we found that the error distribution becomes narrow and shifts toward

62

small CIEDE2000 color differences as the number of terms in the polynomial increases except 6, and 8 terms. Although the ΔE00_avg. of 20 terms is 1.44, its value is not good enough to present the color characteristic of ChLCD. So look-up table with interpolation also would be applied on colorimetric characterization of ChLCD.

At first, four various non-uniform packing methods are chosen, they are cubic of 3, 5, 7, and 9. And then, four different geometrical interpolation methods are used, they are trilinear, prism, pyramid, and tetrahedral. From calculating the ΔE00_avg.

between the predicted values by these 16 various geometrical packing interpolation and measured values of 900 randomized test samples, the results can be summarized below.

When packing number is less, pyramid and tetrahedral interpolation methods are better than another two methods; when packing number is more, there are nearly no differences between these four interpolation methods. And the error amplitude decreases as the number of packing levels increases no matter which the geometrical interpolation method is. Finally, the ΔE_{00_avg.} of these four geometrical interpolation methods in cubic of 9 non-uniform packing are below 1, which means humans cannot recognize these differences. Namely, LUT of non-uniform 9-level packing with various geometrical interpolation techniques successfully make the colorimetric characterization on three-layers-stacked ChLCD. Among these four different interpolation methods, the most beneficial and accurate method is tetrahedral interpolation for our case.

In the near future, no matter what the complex e-Paper or even any display is, these series of colorimetric characterization methods could be completely imitated on

63

it. Then, the best way can be found to be utilized on it.

5.2 Discussions

Although ChLCD has been developed with full color technology, its color performance has a deadly disadvantage which is color shift owing to viewing angle shift. Because its reflective light wavelength obeys Bragg‘s law (λ= n ∙p ∙cosφ , λ:

wavelength of reflective light, n: refractive index, p: pitch of ChLC, φ: incident angle), as φ increases, corresponding λ decreases; as φ decreases, corresponding λ increases.

Fig. 5-1 is a diagram taking for an example to show this phenomenon. When viewing geometry changes from 45^o/0^o to 45^o/+10^o and 45^o/-10^o, it can be taken as 55^o/0^o and 35^o/0^o. The observed light wavelength would change from long wavelength to short wavelength (blue shift) when viewing geometry changes from 45^o/0^o to 45^o/+10^o; from short wavelength to long wavelength (red shift) when viewing geometry changes from 45^o/0^o to 45^o/-10^o. This phenomenon is still opened for future study.

Fig. 5-1 A diagram of ChLCD color shift

64

Reference

1. G. H. Gelinck, et al., ―A rollable, organic electrophoretic QVGA display with field-shielded pixel architecture,‖ JSID 14/2, 113–118 (2006).

2. A. M. Green, et al., ―Energy efficient flexible Reflex™ displays,‖ Proc.

IDRC ‘08, 55–59 (2008).

3. J. Heikenfeld, et al., ―Review Paper: A critical review of the present and future prospects for electronic paper,‖ JSID 19/2, 131 (2011).

4. P. Green, et al., ―Colour Engineering- Achieving Device Independent Colour,‖

John Wiley & Sons, Ltd., 127-129 (2002).

5. J. Morovic, ―Color Gamut Mapping,‖ John Wiley & Sons, Ltd., 45-47 (2008).

6. 胡國瑞等, ―顯示色彩工程學,‖ 全華圖書股份有限公司, 139 (2009).

7. E. F. Schubert, ―Light Emitting Diode,‖ Second Edition, Cambridge University Press, 313-348 (2006)

8. A. A. Afifi, et al., ―Statistical Analysis,‖ New York: Academic Press, 108 (1972) 9. H. R. Kang, ―Color scanner calibration of reflected samples,‖ Proc. SPIE 1670,

468-477 (1992)

10. H. R. Kang, et al., ―Neural network applications to the color scanner and printer calibrations,‖ J. Electron. Imag., Vol. 1: 125-135 (1992)

11. J. M. Kasson, et al., ―A tetrahedral interpolation technique for color space conversion,‖ Proc. SPIE 1909, 127-138 (1993)

12. H. R. Kang, ― Color Technology for Electronic Imaging Devices,‖ SPIE Optical Engineering Press, 67-150 (1997)

13. P. C. Hung, ― Colorimetric calibration in electronic imaging devices using a look-up-table model and interpolations,‖ Journal of Electronic imaging 2(1),

65

53-61 (1993)

14. H. C. Lee, ― Introduction to Color Imaging Science,‖ Cambridge University Press, 401 (2005)

15. X. Y. Huang, et al., ― The transient response and dynamic drive of cholesteric liquid-crystal displays,‖ JSID5/3 , 179 (1997)

16. S. T. Wu, et al., ―Reflective Liquid Crystal Displays,‖ John Wiley & Sons, Ltd., 208 (2001)

17. D. Davis, et al., ―Eight-color high-resolution reflective cholesteric LCDs,‖ SID Intl. Symp. Digest Tech. Papers, 29, 901 (1998)

18. D. K. Yang, et al., ―Control of the reflectivity and bistability in displays using CLC,‖J. Appl. Phys. 76 (2), 1331 (1994)

19. D. Davis, et al., ―Multiple color high resolution reflective cholesteric liquid crystal displays,‖ SID (1997)

20. A. Khan, et al., ―Recent progress in flexible color reflective cholesteric displays,‖

JSID16/2 , 245 (2008)

21. J. P. Allebach, et al., ― Efficient implementation of nonlinear color

transformations,‖ First IS&T/SID Color Imaging Conference: Transforms &

Transportability of Color, 143-148 (1993) 22. http://en.wikipedia.org/wiki/FLEPia

23. http://www.xrite.com/product_overview.aspx?ID=1161