本研究主要目的是探討多原色子畫素演譯黑白圖形影像所產生的邊緣漏色和不平滑程度的現象，利用模擬並配合視覺實驗，提出合理之量化方法以改善影像品質。我們利用矩陣切換演算法做為多原色色彩轉換之依據，完成一電腦程式並利用其功能進行多原色子畫素演譯之影像輸出。並將輸出之影像以人眼視覺模型做模糊化後計算各方向之影像黑白邊緣處色彩的chroma對徑向空間積分值，以各方向的積分結果平均值作為該影像邊緣漏色之量化依據。

邊緣不平滑程度則使用經人眼視覺模糊化後的影像中L*=50的等亮度線於極座標中之傅立葉係數加總作為量化計算方式，分別對9種不同六色子畫素排列所演譯之黑白橢圓形影像之邊緣漏色和不平滑程度進行量化。　

視覺實驗則使33位受測者對9種畫素排列所演譯之影像之邊緣漏色與不平滑程度給分，最後進行所提的模擬量化方法與視覺實驗比較，結果顯示邊緣漏色模擬和實驗結果之Spearman相關係數為0.56，而邊緣平滑化的模擬和實驗結果有0.91的相關性，結果顯示對於影像的平滑程度量化的演算法適用於預測人眼視覺的程度，而邊緣漏色的量化演算法或視覺實驗方法則有待改進。

In this thesis, we mainly investigate the phenomena of pseudo color edges and jaggy boundaries occurred in a black-white rendering elliptic image with six-primary color sub-pixels. By means of simulation and human visual experiments, we propose a reasonably quantified way to guide how to improve imaging performance. Using the matrix switching algorithm, we completed a program to execute color conversion from three to six primary color systems, and render a few six-primary color images of black-white ellipses on a liquid crystal display for human visual experiments. In addition, considering low pass filter effect of human vision, we calculated the radial integral values of chroma with respect to all azimuthal directions which azimuthal average values were used to quantitatively evaluate the degrees of pseudo color edges. Similarly, the degrees of jaggy boundaries were evaluated with an iso-luminance contour at L*=50, which was quantified with Fourier analysis in the polar coordinate. For human visual experiments, the imaging characteristics of 9 typical rendering images were evaluated by 33 observers including the degrees of pseudo color edges and jaggy boundaries. From the comparisons of simulation and human visual experiments, they shows the Spearman coefficient is 0.56 in the evaluation of pseudo color edges, and it is 0.91 in the evaluation of jaggy boundaries. It seems that our proposed way is more correlative in the evaluating degrees of jagged boundaries than those in the evaluating degrees of pseudo color edges.

[1]M. R. Pointer, “The gamut of real surface colours,” Color Research and Application, Vol. 5, No. 3, pp.145-155 (1980).
[2] E. Chino et al., “Development of Wide-Color-Gamut Mobile Displays with Four-primary-color LCDs”, SID 06 Digest, pp.1221-1224 (2006)
[3]C. C. Chan et al., “Development of multi-primary color LCD”, IDMC poster, 3-6 July 2007, Taipei, Taiwan (網路資料)
[4] Masatsugu Teragawa “Four Primary Color (RGB+Y) Display : The Latest Technologies and Advantages”, IDW, pp.1353-1356, (2010)
[5] Y. C. Yang et al., “Development of six primary-color LCD,” SID 05 Digest, pp.1210-1213 (2005).
[6]T. Ajito et al., “Color conversion method for multiprimary display using matrix switching,” Optical Review, Vol. 8, No. 3, pp.191-197 (2001).
[7] Nonaka, T. et al., “Color mixing simulator for display surfaces based on human color vision ,” Proceedings 21st European Conference on Modeling and Simulation , ECMS , pp.304-308.(2007)
[8] Wonhee C. et al.,"Subpixel Rendering Method for Color Error Minimization on Subpixel Structured Display",SPIE-IS&T/ Vol. 6058, pp.60580F(1-9) (2006) [9]http://en.wikipedia.org/wiki/Cone_cell
[10]周世生, “印刷色彩學” 北京:印刷工業出版社 p.39 (2005)
[11]周世生, “印刷色彩學” 北京:印刷工業出版社 p.13 (2005)
[12]http://en.wikipedia.org/wiki/CIE1931
[13]http://climatesanity.wordpress.com/2009/07/18/more-on-compact-fluorescent-lights/
[14] http://www.associatepublisher.com/e/c/co/color_temperature.htm
[15]http://www.ecse.rpi.edu/~schubert/Light-Emitting-Diodes-dot-org/chap17/chap17.htm
[16] http://en.wikipedia.org/wiki/Macadam_Ellipse
[17] Nonaka, T. et al., “Analytical Simulator of Color Mixing Process on Large Display Surfaces” ,Systems, Man and Cybernetics, ISIC. IEEE International Conference. (2007)
[18] Stephen Westland and Caterina Ripamonti “Computational Colour Science using MATLAB” Ch7.,1st edition ,John Wiley & Sons.(2004)
[19] H. Motomura, “Color conversion for a multi-primary display using linear interpolation on equi-luminance plane method (LIQUID)”, Journal of the SID, Vol. 11, Issue 2, pp.371-378 (2003).
[20] http://en.wikipedia.org/wiki/Spearman_correlation
[21] Yu-Kuo Cheng and Han-Ping D. Shieh , “Relative Contrast Sensitivity for Color Break-Up Evaluation in Field-Sequential-Color LCDs”, Journal of display technology ,vol. 5, no. 10 , pp.379-384 ,(2009)