近年來，薄膜電晶體液晶顯示器(TFT-LCD)已逐漸成為平面顯示器（FPD）的主流，因此TFT-LCD的影像品質也愈來愈受重視。而在TFT-LCD的影像品質分析上，有些現象是很難用一個相當確定的標準加以評估，例如mura、殘影等。而在TFT-LCD中，mura是相當普遍的缺陷。目前TFT-LCD製造商對於mura的鑑定都是以有經驗、技巧的檢測人員來評估。不過很明顯的，藉由人工的檢測勢必成本會提高且評估的結果會較無一致性。所以為了要克服這些缺點，自動化量測分析mura的系統勢必是需要的。
為了要建立一套自動化量測mura的光學系統，所以本論文試圖利用人眼視覺的感覺來訂定評估mura的標準。在針對mura的實驗方面，我們首先利用邊界清晰的mura發現恰辨差對比(Cjnd)與背景亮度、mura大小及mura形狀等參數有關。而隨著背景亮度增加，恰辨差對比也會隨之下降，不過當背景亮度增加到一個臨界值(約172 cd/m2)以上時，其恰辨差對比則沒有太大的變化，所以對於人眼的感覺我們發現恰辨差對比與背景亮度呈非線性關係。而較小面積的mura其恰辨差對比也會比較高，其中線狀mura的恰辨差對比均較其他形狀mura的來得高。另外藉由漸進式mura的實驗，我們可以找出估計mura的人眼視覺臨界值，而利用此臨界值與邊界清晰mura的結果相較下，也從中發現人眼在檢測任何型態的mura都是以恰辨差的準則去評斷。
基於上述結果的分析，我們得到一個較符合人眼感覺的恰辨差對比稱為MCjnd，同時也建議此MCjnd方程式能夠替換SEMU定義的Cjnd。而此新定義的Mura標準我們稱之為MSEMU。所以經由比較MSEMU與SEMU的分析結果，可證實MSEMU確實較符合人眼視覺的感覺。

In recent years, the TFT-LCD gradually becomes a mainstream of FPDs (Flat Panel Displays), and the request of image quality for a TFT-LCD becomes more and more severe. There are many items such as luminance, chromaticity, contrast, flicker, crosstalk and response time, etc. have to be evaluated for the image quality of display devices. But several of the other items have difficulties during evaluation of image quality, e.g. mura and image sticking. Mura is the most popular defect in producing TFT-LCD, in most cases, and is not easily identified so that those persons identifying mura in the industry need the experienced skill based on the related knowledge. It is thus obvious for manually identifying process by human beings to be costly and inconsistent. To overcome such hardship, an automated process for mura analysis has been considered.
In this study, we have investigated, analyzed and quantified the mura phenomena in the image quality of a TFT-LCD and hopefully try to set-up an optical evaluation system for mura. We found that the Cjnd (Just-Noticeable-Difference Contrast) value of the mura with abrupt boundary will be influenced by background luminance, mura size, and mura shape. As the background luminance increased, the Cjnd value decreased. However, when the background luminance was increased to above a threshold value around 172 cd/m2, there was no significant difference in Cjnd to human visual perception for a given mura size. Obviously, the relationship between Cjnd and background luminance was non-linear. Furthermore, the Cjnd value of mura with a smaller size was higher, and the Cjnd value of line type mura was higher than those of the other muras which with the same size but different shape. Besides, the threshold value for human visual perception of mura could be estimated with a graded mura. Comparing the results of Muras with abrupt boundary and graded ones, we also found that any type of mura detectable by human eyes could be estimated by the index of Just Noticeable Difference contrast.
Based on the analysis of our results, we modified the Cjnd equation which proposed by SEMI (Semiconductor Equipment and Materials International), and re-named as MCjnd which was more correlated to human visual inspection. Then, replacing the Cjnd with MCjnd, a new definition of mura level was obtained and called MSEMU, in contrast to SEMU proposed by SEMI. After comparing the analysis results of MSEMU and SEMU, it could be concluded that the MSEMU definition was reasonable and more correlated to human visual inspection for the mura analysis.

[1] www.auo.com.
[2] 紀國鐘、鄭晃忠主編，液晶顯示器技術手冊，台灣電子材料與元件協會出版，經濟部技術處發行，91年版，ISBN：9572812505.
[3] 趙中興， “顯示器的基本原理與製造技術”，全華書局， 89年版，ISBN：9572127241.
[4] 顧鴻壽，“光電液晶平面顯示器技術基礎及應用”，新文京開發出版有限公司，ISBN：9575129571.
[5] Y. Mori, T. Tamura, R. Yoshitake, K. Moriguchi, K. Tanahashi and S. Tsuji, “Ergonomics Approach to Evaluate Minimum Perceivable Non-uniformity in Liquid Crystal Displays,” IDW, VHFp-2, pp. 1315~1318, 2002.
[6] New Standard: Definition of Measurement and Index (SEMU) for Luminance Mura in FPD Image Quality Inspection, SEMI Draft Document #3324.
[7] T. Tamura, M. Baba and T. Furuhata, “Effect of the Background Luminance on Just Noticeable Difference (JND) Contrast of “Mura” in LCDs,” IDW, VHFp-4, pp. 1527~1530, 2002.
[8] Y. Mori, T. Tamura, R. Yoshitake, K. Moriguchi and S. Tsuji, “Evaluation of Luminance Non-uniformity “Mura” in Liquid Crystal Displays by Observations of Just Noticeable Differences,” IDW, VHFp-4, pp. 1685~1688, 2001.
[9] Don Gyou Lee, Il Ho Kim, Mun Cheol Jeong, Baek Keun Oh and Woo Yeol Kim, “Mura Analysis method by using JND Luminance and the SEMU Definition,” IDW, VHF1-2, pp. 1467~1470, 2003.
[10] Y. Mori, K. Tanahashi, R. Yoshitake, T. Tamura, K. Moriguchi, T. Yoshizawa and S. Tsuji, “Measurement System and Detection Method of “Mura” in TFT-LCD,” IDMC 2003, Th 14.05.
[11] Stevens, S.S., Psychophysics. Introduction to its Perceptual, Neural, and Social prospects. New York: John Wiley and Sons., 1975.
[12] G. Wyszecki and W. S. Stiles, “Color Science Concepts and Method, Quantitative Data and Formulae”, John Wiley and Sons., 1982.
[13] 李江山等人, “視覺與認知” 遠流書局, ISBN: 957-32-3669-9.
[14] http://202.120.156.20/2005/courseware/home.htm.
[15] http://genpsy.dlearn.kmu.edu.tw/sens/Sensation2.html.
[16] http://bbs.ee.ntu.edu.tw/boards/Programming/14/3.html.
[17] http://www.socialwork.com.hk/psychtheory/theory_psy/.
[18] Ming-Cong Hong, “Muti-Resolution and JND-Based Image Segmentation Algorithm,” M. S. Thesis, NYUST, Taiwan, R.O.C., 2000.
[19] K. Moriguchi, R. Yoshitake and Y. Mori, “An Ergonomic Approach to Evaluate Screen Qualities on Electronic Visual Displays,” IDW, VHF1-3, pp. 1471~1474, 2003.
[20] S.Hecht, “The Visual Discrimination of Intensity and The Weber-Fechner Law,” The Journal of General Physiology, pp. 235~267, 1924.
[21] Y. Mori, K. Tanahashi and S. Tsuji, “Quantitative Evaluation of Visual Performance of Liquid Crystal Displays,” for Optical Information Processing (The International Society for Optical Engineering), Vol.4113, pp.242-249, 2000.
[22] R. Yoshitake, T. Tamura and S. Tsuji, “A Proposal for a Quantitative Model of “Mura” Level of LCDs on the Basic of Human Senses”, The journal ITE, Vol.56, No.7, pp.1153~1158, 2002.
[23] Y. Mori, T. Tamura, R. Yoshitake, K. Moriguchi, K. Tanahashi and S. Tsuji, “Quantitative Methods of Luminance Non-Uniformity in Liquid Crystal Displays Based on Just Noticeable Difference”, The journal ITE, Vol.56, No.11, pp.1837~1840, 2002.