在印刷現場影響印刷色彩表現的因素有兩個，分別是印刷油墨和光源。印刷油墨配方與色彩特性之間的對應關係一直是印刷色彩複製領域中備受重視的研究項目，目前已發展出基於多色網點面積率的數學模型來預測印刷油墨配方的色彩再現性，例如，諾克伯光譜模型為基於原色油墨的光譜反射率及其網點面積率建立的半色調印刷色彩預測模型，但僅以簡單的線性方程式來描述原色光譜反射率及網點面積率之間的關係是不夠精確的。另外，印刷工業使用的標準光源為色溫 5000 K 的高演色性光源，目前市面上有照明指示貼紙可以用來檢測印刷現場的彩色打樣用光源是否接近 CIE 標準光源，其設計原理即是利用條件等色性，當照明環境越符合指示貼紙所指示的日光光源時，貼紙上的兩個色塊所呈現的顏色越接近。在LED 新式光源發明之後，因為其壽命長、節能、高發光效率等等優勢，逐漸取代耗能的傳統光源（例如：螢光燈、鎢絲燈），由於 目前作為主流的藍/黃二色型白光LED 光源光譜能量分布較不均勻，與CIE D50日光光源的光譜能量分布差異非常大，導致照明指示貼紙在 5000 K 白光 LED 光源下無法發揮功能。
本論文根據上述背景設計了以下三項實驗，分別為：(1) 實驗一「優化諾克伯印刷色彩預測模型」，在印刷油墨的控制上，針對諾克伯印刷色彩預測模型進行優化並評估優化後模型的色彩預測能力；(2) 實驗二「設計同色異譜對光譜：考量LED與日光」，為利用同色異譜設計照明檢測導具，目的在於區分 5000 K白光LED照明與 CIE D50日光照明；(3) 實驗三「設計同色異譜對光譜：考量高/低演色性LED」，同樣利用同色異譜設計可實際應用於印刷現場的照明檢測導具，目的在於區分高演色性 5000 K 白光 LED 照明低演色性 5000 K 白光 LED 照明。
實驗一探討4種諾克伯印刷色彩預測模型之優化分析，包含「諾克伯光譜模型」、「修正諾克伯光譜模型」、「修正諾克伯網點面積率模型」與「修正網點面積率與諾克伯光譜模型」，實驗中先製作及量測實驗樣本，對諾克伯印刷色彩預測模型進行訓練，取得各模型優化係數後測試優化之模型，並計算色差來分析優化後模型之預測色彩能力。實驗二為利用同色異譜現象設計符合「在5000 K白光LED照明環境下無色差，在標準D50照明環境下有色差」條件之同色異譜對光譜反射率，利用主成分分析法解構指定之色彩資料庫，並以ColorChecker導表上的18個有彩度色塊作為參考色，對其進行光譜反射率預測，接著設定複數篩選條件來篩選出符合實驗目的之同色異譜對。實驗三為利用同色異譜現象設計符合「在低演色性白光LED照明下無色差，在高演色性白光LED照明下有色差」條件及「在高演色性白光LED照明下無色差，在低演色性白光LED照明下有色差」條件之同色異譜對光譜反射率。
實驗一結果顯示，優化後的色彩預測模型以修正網點面積率與諾克伯光譜模型的預測色彩能力最佳。實驗二成功產生「在5000 K白光LED照明環境下無色差，在標準D50照明環境下有色差」同色異譜對之光譜反射率。實驗三結果成功產生「在低演色性白光LED照明下無色差，在高演色性白光LED照明下有色差」及「在高演色性白光LED照明下無色差，在低演色性白光LED照明下有色差」同色異譜對之光譜反射率。由本論文提出之光譜預測法及篩選條件，可有效模擬在指定白光LED光源下達到等色之同色異譜對。

Printing ink amounts and light sources are two important factors that affect color appearances of printed media. Color prediction model between CMY signal amounts and ink reflection spectra is one of the attractive researches in the field of printing color reproduction for a long time. Several mathematical models have been developed to predict color performances of halftone colors, such as the spectral Neugebauer model, which was developed for predicting printing color based on the reflectance spectra and the dot area of the Neugebauer primary colors. The basic spectral Neugubauer model which described the relationship between CMY signal amounts and ink reflection spectra using a linear equation was completely inaccurate. In addition, Metamerism is a phenomenon that two objects or colors with different spectra show similar colors under a particular light source, while showing different colors under the other sources. There are several types of paper-based lighting indicators which make use of metamerism property are released for detecting standard illuminants currently. It has been found that the D50 lighting indicator cannot work well for general white-light LEDs.
Based on the above motivation, this study designed three experiments as follows: (1) Experiment 1 named “Optimized the Neugebauer color prediction model” is designed to optimize the CMY dot areas, several Neugebauer-based color prediction models were assessed according to color predicting abilities; (2) Experiment 2 named “Spectral design of metamer in terms of LED and daylight”. It designed a novel color quality inspection guide for distinguishing between white-light LED and CIE daylight based on metamerism; (3) Experiment 3 named “Spectral design of metamer in terms of high and low color rendering LEDs”. It designed a novel color quality inspection guide for distinguishing between the white-light LED with low color rendering and the white-light LED with high color rendering.
Experiment 1 discussed the optimization analysis of four color prediction models, including the spectral Neugebauer model (SN model), modified spectral Neugebauer model (m-SN model), modified dot-coverage Neugebauer model (m-DN model), and modified dot-coverage and spectral Neugebauer model (m-DSN model). The training color samples are used for training the color prediction models to obtain the compensation coefficients of each model, and the testing samples are used for testing the color prediction models. In Experiment 2, the reflectance spectra of metamer with the capabilities of showing similar colors under 5000 K white-light LED were designed, while showing different colors under CIE daylight series. The spectral prediction methods were making use of principal component analysis (PCA) to rebuild the reflectance spectra of 18 color patches chosen from ColorChecker Chart. Experiment 3 designed two kinds of metamer: One has the capability of showing similar colors under white-light LED with low color rendering, while showing different colors under white-light LED with high color rendering. The other one has the capability of showing similar colors under white-light LED with high color rendering, while showing different colors under white-light LED with low color rendering.
In Experiment 1, the optimized m-DSN model with the best ability to predict printing colors was confirmed. In Experiment 2, we successfully developed the metameric color pairs which have the capabilities of showing similar color under 5000 K white-light LED. In Experiment 3, we also successfully developed two kinds of metameric color pairs, one pair have the capability of showing similar color under white-light LED with low color rendering, while showing different colors under white-light LED with high color rendering. The other pair has the capability of showing similar color under white-light LED with high color rendering, while showing different colors under white-light LED with low color rendering. Consequently, the experimental results showed that the proposed spectrum prediction method can effectively simulate the metameric pairs under a specified light source.

[1] M. Hébert, "Optical models for color reproduction," PhD, Institut d’Optique Graduate School, 2013.
[2] R. S. Berns, Billmeyer and Saltzman's principles of color technology, 3rd ed. John Wiley & Sons, 2000.
[3] H. T. Gu, M. R. Pointer, X. Y. Liu, and M. R. Luo, "Quantifying the suitability of CIE D50 and a simulators based on LED light sources," Color Res. Appl., vol. 42, no. 4, pp. 408-418, 2017.
[4] C. S. McCamy, "Metamers for assessing the quality of CIE D50 simulators," Color Res. Appl., vol. 21, no. 3, pp. 236-238, 1996.
[5] M. Vik, L. Hes, and M. Viková, "Chromatic Adaptation and Metameric Pair Evaluation," 2006.
[6] X. Lv and M. R. Luo, "LED Simulators for the Reproduction of the New CIE Standard LED Sources," in Proceedings of the 29th CIE Session, Washington D.C., USA, June 14-22, 2019 2019, pp. 93-100, doi: 10.25039/x46.2019.OP16.
[7] 胡國瑞、孫沛立、徐道義、陳鴻興、黃日鋒、詹文鑫、羅梅君　編著，「顯示色彩工程學（第二版）」，台北，全華圖書股份有限公司，2011。
[8] 羅梅君　編著，「數位色彩管理科學：色彩度量學」，藍海文化，2010。
[9] H. Xu, M. R. Luo, and B. Rigg, "Evaluation of daylight simulators. Part 2; Assessment of the quality of daylight simulators using actual metameric pairs," Color. Technol., vol. 119, no. 5, pp. 253-263, 2003, doi: 10.1111/j.1478-4408.2003.tb00180.x.
[10] R. Balasubramanian, "Optimization of the spectral Neugebauer model for printer characterization," J. Electron. Imaging, vol. 8, no. 2, pp. 156-167, 1999.
[11] J. A. C. Yule and W. J. Nielsen, "The penetration of light into paper and its effect on halftone reproduction," in Proc. TAGA, 1951, vol. 3, pp. 65-76.
[12] R. D. Hersch and F. Crété, "Improving the Yule-Nielsen modified Neugebauer model by dot surface coverages depending on the ink superposition conditions," in Color Imaging X: Processing, Hardcopy, and Applications, 2005, vol. 5667: International Society for Optics and Photonics, pp. 434-447.
[13] D. H. Foster, K. Amano, S. M. C. Nascimento, and M. J. Foster, "Frequency of metamerism in natural scenes," J. Opt. Soc. Am. A, vol. 23, no. 10, pp. 2359-2372, 2006, doi: 10.1364/JOSAA.23.002359.
[14] A. Akbarinia and K. R. Gegenfurtner, "Color metamerism and the structure of illuminant space," J. Opt. Soc. Am. A, vol. 35, no. 4, pp. B231-B238, 2018.
[15] W. G. Kuo and M. R. Luo, "Methods for quantifying metamerism. Part 1 - Visual assessment," J. Soc. Dye. Colour., vol. 112, no. 11, pp. 312-320, 1996.
[16] J. B. Cohen and W. E. Kappauf, "Metameric color stimuli, fundamental metamers, and Wyszecki's metameric blacks," Am. J. Psychol., pp. 537-564, 1982.
[17] G. W. Wyszecki, "Evaluation of metameric colors," J. Opt. Soc. Am., vol. 48, no. 7, pp. 451-454, 1958.
[18] G. W. Wyszecki and W. S. Stiles, Color Science: Concepts and Methods, Quantitative Data and Formulae, 2nd ed. John Wiley and sons, New York, 1982.
[19] K. Takahama and Y. Nayatani, "New method for generating metameric stimuli of object colors," J. Opt. Soc. Am., vol. 62, no. 12, pp. 1516-1520, 1972.
[20] W. S. Stiles and G. W. Wyszecki, "Counting metameric object colors," J. Opt. Soc. Am., vol. 52, no. 3, pp. 313-328, 1962.
[21] W. S. Stiles, G. W. Wyszecki, and N. Ohta, "Counting metameric object-color stimuli using frequency-limited spectral reflectance functions," J. Opt. Soc. Am., vol. 67, no. 6, pp. 779-784, 1977.
[22] N. Ohta, "A simplified method for formulating pseudo‐object colors," Color Res. Appl., vol. 7, no. 2, pp. 78-81, 1982.
[23] S. A. Burns, J. B. Cohen, and E. N. Kuznetsov, "Multiple metamers: preserving color matches under diverse illuminants," Color Res. Appl., vol. 14, no. 1, pp. 16-22, 1989.
[24] G. D. Finlayson and P. Morovic, "Metamer sets," J. Opt. Soc. Am. A, vol. 22, no. 5, pp. 810-819, 2005.
[25] H. S. Fairman and M. H. Brill, "The principal components of reflectances," Color Res. Appl., vol. 29, no. 2, pp. 104-110, 2004.
[26] W. S. Stiles and G. W. Wyszecki, "Intersections of the spectral reflectance curves of metameric object colors," J. Opt. Soc. Am., vol. 58, no. 1, pp. 32-40, 1968.
[27] N. Ohta and G. W. Wyszecki, "Location of the nodes of metameric color stimuli," Color Res. Appl., vol. 2, no. 4, pp. 183-186, 1977.
[28] N. Ohta, "Intersections of spectral curves of metameric colors," Color Res. Appl., vol. 12, no. 2, pp. 85-87, 1987.
[29] S. Westland, C. Ripamonti, and V. Cheung, Computational colour science using MATLAB, 2nd ed. John Wiley & Sons, 2012.
[30] CIE, 015: 2018 Colorimetry. Vienna: Commission Intermationale de I’Eclairage, 2018.