基于视觉匹配的LCD特性化方法研究
|
摘要
随着近年来数字技术、彩色数字图像的不断发展和广泛使用,颜色在不同的设备之间精确地传递或再现成为该领域的重要课题,色彩管理系统(ColorManagement System)的任务就是为了实现这一目的。在颜色管理系统中,最基本的工作就是对输入/输出设备进行特性化。
彩色TFT-LCD是一种目前市场上流行的液晶显示器,也是未来显示器的发展方向之一。随着其成本的逐步降低,质量逐步提高,大有取代CRT的趋势。
论文以LCD特性化方法为主要研究内容,在广泛查阅国内外文献的基础上,采用基于视觉匹配的设备特性化方法,同时完成了孟塞尔色卡在数码相机和LCD之间的复制。具体完成的研究工作包括:
1.通过视觉匹配方法获取CIEXYZ色空间到LCD的RGB空间的训练样本对。将一个给定观察条件下的一些色貌因素“映射”到另一媒体上。
2.用视觉匹配结合BP神经网络方法建立LCD的XYZ(?)RGB的非线性映射关系。选取的训练样本数量较少,得出了数码相机特性化平均△E~*ab为4.2761;最大△E~*ab为11.1962;最小△E~*ab为0.9713。LCD特性化平均△E~*ab为4.2711;最大△E~*ab为12.1545;最小△E~*ab为0.4108。这个色差属于可接受范围,可以满足多数实际应用的要求。
本文的研究成果不仅对彩色数字输入输出设备技术有理论价值,而且在印刷、医疗诊断、化妆品生产、各种照明条件下不同媒体颜色再现等领域有应用前景。
With the development and wide application of color imaging during recent years,the communication and reproduction of color information among different devices is becoming an important subject and is the task of Color Management System(CMS).In the CMS,the main works is to character the input/output devices.
The color TFT-LCD is a kind of popular LCDs in present market,and is a development orientation of the future LCDs.It tends to replace CRT owing to the decreasing cost and increasing quality.
This thesis concentrates on the characterization of LCD.After reading a lot of references,we provide a method of device characterization based on visionmatching and accomplish the reproduction of Munsell Color Chips between still digital camera and LCD.The corresponding research work include the following aspects:
1.After studying the properties of the present characterization methods.We obtain the training sample couple—CIEXYZ color space and RGB space of LCD via "visualmatching" method,and mapped some color appearance factors to another media under the certain observational conditions.
2.we built the nolinear mapping relation between XYZ and RGB of LCD,carried out LCD characterization by virtue of the Back-Propagation(BP) artificial neutral network and "Visualmatching".Andwe chose and studied a few samples:the mean of digital camera's characterization,ΔE~*ab,is 4.2761;the maximalΔE~*ab is 11.1962;the minimalΔE~*ab is 0.9713;the mean of LCD's characterization,ΔE~*ab,is 4.2711;the maximalΔE~*ab is 12.1545;the minimalΔE~*ab is 0.4108.These color differences are acceptable and can satisfy most of actual application requirements.
The research results in this paper have not only some theoretical importance, but have good application foreground in such fields as print,medical diagnoses, cosmetic production and the reproduction of different media colors under various lighting conditions about the face color.
彩色TFT-LCD是一种目前市场上流行的液晶显示器,也是未来显示器的发展方向之一。随着其成本的逐步降低,质量逐步提高,大有取代CRT的趋势。
论文以LCD特性化方法为主要研究内容,在广泛查阅国内外文献的基础上,采用基于视觉匹配的设备特性化方法,同时完成了孟塞尔色卡在数码相机和LCD之间的复制。具体完成的研究工作包括:
1.通过视觉匹配方法获取CIEXYZ色空间到LCD的RGB空间的训练样本对。将一个给定观察条件下的一些色貌因素“映射”到另一媒体上。
2.用视觉匹配结合BP神经网络方法建立LCD的XYZ(?)RGB的非线性映射关系。选取的训练样本数量较少,得出了数码相机特性化平均△E~*ab为4.2761;最大△E~*ab为11.1962;最小△E~*ab为0.9713。LCD特性化平均△E~*ab为4.2711;最大△E~*ab为12.1545;最小△E~*ab为0.4108。这个色差属于可接受范围,可以满足多数实际应用的要求。
本文的研究成果不仅对彩色数字输入输出设备技术有理论价值,而且在印刷、医疗诊断、化妆品生产、各种照明条件下不同媒体颜色再现等领域有应用前景。
With the development and wide application of color imaging during recent years,the communication and reproduction of color information among different devices is becoming an important subject and is the task of Color Management System(CMS).In the CMS,the main works is to character the input/output devices.
The color TFT-LCD is a kind of popular LCDs in present market,and is a development orientation of the future LCDs.It tends to replace CRT owing to the decreasing cost and increasing quality.
This thesis concentrates on the characterization of LCD.After reading a lot of references,we provide a method of device characterization based on visionmatching and accomplish the reproduction of Munsell Color Chips between still digital camera and LCD.The corresponding research work include the following aspects:
1.After studying the properties of the present characterization methods.We obtain the training sample couple—CIEXYZ color space and RGB space of LCD via "visualmatching" method,and mapped some color appearance factors to another media under the certain observational conditions.
2.we built the nolinear mapping relation between XYZ and RGB of LCD,carried out LCD characterization by virtue of the Back-Propagation(BP) artificial neutral network and "Visualmatching".Andwe chose and studied a few samples:the mean of digital camera's characterization,ΔE~*ab,is 4.2761;the maximalΔE~*ab is 11.1962;the minimalΔE~*ab is 0.9713;the mean of LCD's characterization,ΔE~*ab,is 4.2711;the maximalΔE~*ab is 12.1545;the minimalΔE~*ab is 0.4108.These color differences are acceptable and can satisfy most of actual application requirements.
The research results in this paper have not only some theoretical importance, but have good application foreground in such fields as print,medical diagnoses, cosmetic production and the reproduction of different media colors under various lighting conditions about the face color.
引文
[1]J.Morovic,M.R.Luo.The fundamentals of gamut mapping:a survey[J].Journal of Imaging Science and Technology,2000,45(3):283-290.
[2]C.Lee,Y.Park,et al.Gamut mapping algorithm using lightness mapping and multiple anchor points for linear tone and maximum chroma reproduction[J].Journal of Imaging Science and Technology,2001,45(3):209-223.
[3]黄庆梅,彩色图像复制及色域映射技术的研究[D].北京理工大学,2002.
[4]黄庆梅,赵达尊.不同色域映射算法视觉效果的一种评价方法[J].北京理工大学学报:2003,23(4):408-413.
[5]王勇,数字图像设备颜色特征化及其再现理论与技术研究[D].浙江大学,2006.
[6]w.b.Cowan.An Inexpensive scheme of calibration of a colour monitor in terms of CIE standard coordinates[J].Computer Graphics,1983,17(3):315-321.
[7]R.S.Berns.Methods for characterizing CRT displays[J].Displays,1996,16(4):173-182.
[8]R.Bala.Reducing the cost of lookup table based color transformations[J].Journal of Imaging Science and Technology,2000,44(4):321-327.
[9]廖宁放,杨卫平,等.用多隐层BP网实现的CRT色度变换[J].中国图象图形学报.2000,5(6):470-473.
[10]胡涛.计算机CRT的色彩管理研究[J].计算机工程与应用,2002,(24):101-103.
[11]W.Yang,N.Liao,et al.color reproduction on CRT displays via BP neural networks under office environment[J].Journal of Beijing Institute of Technology,2003,12(4):376-380.
[12]N.Liao,Z.Gao.A comparative study of a CRT colorimetric prediction model by neural networks and the models by conventional method[J].Color Research and Application,1999,24(1):45-51.
[13]N.Tamura,N.Tsumura,et,al.masking Model for accurate colorimetric characterization of LCD[J].2003,Journal of the SID,11(2):1-7.
[14]G.Sharma.Comparative evaluation of color characterization and gamut of LCDs verus CRTs[C].2002,SPIE,4663:177-186.
[15]Y.Kwak,L.Macdonald.Characterisation of desktop LCD projector[J].Displays,2000,21(5):179-194.
[16]M.D.Fairchild,D.R.Wyble.Colorimetric characterization of the apple studio display(flat panel LCD)[R].Munsell Color Science Laboratory,1998.
[17]黄庆梅,赵达尊,等.LCD液晶显示器的颜色特性研究[J].光学技术,2005,31(2):243-245.
[18]T.Johnson.Methods for characterizing colour scanners and digital cameras.displays[J].1996,16(4):183-191.
[19]Y.Wang,H.Xu.Spectral characterization of a flat panel color scanner using PCA method[C].SPIE,2006,6033,60330J-1-5.
[20]C.Lee,E.Lee,et al.Color space conversion via gamut-based colog samples of printer[J].Journal of Imaging Science and Technology,2001,45(5):427-435.
[21]T.Johnson.Methods for characterizing colour printers[J].Displays,1996,16(4):193-202.
[22]黄庆梅,赵达尊.人工神经网络作为打印机色空间转换工具的问题[J].光学技术,2003,29(2):146-148.
[23]胡威捷,汤顺清,朱正芳.现代颜色技术原理及应用.北京:北京理工大学出版社,200731-97.
[24]王春锴.用于TFT-LCD驱动控制系统的电源管理单元的研究与设计[D].上海:复旦大学,2008.
[25]翁星光.基于LCD显示特性文件质量评测研究[D].江南大学,2007.
[26]唐剑.TFT-LCD Mura缺陷检测研究[D].北京交通大学,2008.
[27]钟云飞,陈乾坤.彩色LCD显示器色彩特性研究[J].湖南工业大学学报,2008,22(1):96-98.
[28]杨健,郝飞.显示器视觉特性化研究[J].铜仁学院学报,2008,10(5):120-122.
[29]Yasuhiro Yoshida,Yoichi Yamamoto.Color Calibration of LCDs[C].The Seventh Color Imaging Conference,2002,305-311.
[30]Nobuhiko Tamura,Norimichi Tsumura and Yoichi Miyake.Maksing Model for Accurate Colorimetric Characterization of LCD[C].IS&T/SID Tenth Color Imaging Conference,2002,312-316.
[31]Ellen A.Day,Lawrence Taplin and Roy S.Berns.Colorimetric Characterization of a Computer-Controlled Liquid Crystal Display[J].Color Res.& Appl,2004,29(5):365-373.
[32]Wen Senfar,Wu Royce.Two-primary crosstalk model for characterizing liquid crystal displays[J].Color Research and Application,2006,30(2):102-108.
[33]张显斗,徐海松.液晶显示器颜色特征化的分段分空间模型[J].光学学报,2007,27(9):1719-1724.
[34]杨卫平,廖宁放,等.基于孟塞尔系统的阴极射线管特性化新方法[J].光学学报,2004,24(8):1039-1044.
[35]Simon F T,Frost J A.A new method for the conversion of CIE colorimetric data to munsell notations[J].Color Researchand Application,1987,12(5):256-260.
[36]Tominaga S.Color notation conversion by neural networks[J].Color Research and Application,1993,18(5):253-259.
[37]Shiro Usui,Shigeki Nakauchi,et al.Reconstruction of munsell color space by a five-layer neural network[J].J.Opt.Soc.Am.(A),1992,9(4):516-520.
[38]Wyble D R,Fairchild M D.Prediction of munsell appearance scales using various color-appearance models[J].Color Research and Application,2000,25(2):132-144.
[39]T.L.V.Cheung and S.Westland,Color Camera Characterisation Using Artificial Neural Networks[R],Colour & Imaging Institute,University of Derby,Kingsway House Kingsway,Derby,United Kingdom,IS & T/SID Tenth Color Imaging Conference.
[40]Zhou.S.Q.Zhao.D.Z,BP network based color control in color printers[J],Optical technique,2000,1.
[41]Yu.H.F.Li.C.Y.Shi.J.S,Study of Color Reproduction and Color Prediction from Scanner Image to Printer Image Using Back-Propagation Neutral Networks[J],Proceedings of SPIE, 2002,4922.
[42]杨卫平,跨媒体颜色复制技术简化研究[D],北京理工大学,2005.
[43]CIE TC8-04.Chromatic adaptation under mixed illumination condition when comparing softcopy and hardcopoy images.2004.Vienna,AUSTRIA:COMMISSION INTERNATIONALE DE L'ECLAIRAGE CIE Central Bureau,ISBN 3901906xxy,http://www.cie.co.at/.
[44]ROY S.Berns.颜色技术原理[M].北京:化学工业出版社材料科学与工程出版中心,2002.
[45]束越新.颜色光学基础理论[M].济南:山东科学技术出版社,1981.
[46]大田 登著,刘中本译.色彩工学[M].西安:西安交通大学出版社,1997,173.
[2]C.Lee,Y.Park,et al.Gamut mapping algorithm using lightness mapping and multiple anchor points for linear tone and maximum chroma reproduction[J].Journal of Imaging Science and Technology,2001,45(3):209-223.
[3]黄庆梅,彩色图像复制及色域映射技术的研究[D].北京理工大学,2002.
[4]黄庆梅,赵达尊.不同色域映射算法视觉效果的一种评价方法[J].北京理工大学学报:2003,23(4):408-413.
[5]王勇,数字图像设备颜色特征化及其再现理论与技术研究[D].浙江大学,2006.
[6]w.b.Cowan.An Inexpensive scheme of calibration of a colour monitor in terms of CIE standard coordinates[J].Computer Graphics,1983,17(3):315-321.
[7]R.S.Berns.Methods for characterizing CRT displays[J].Displays,1996,16(4):173-182.
[8]R.Bala.Reducing the cost of lookup table based color transformations[J].Journal of Imaging Science and Technology,2000,44(4):321-327.
[9]廖宁放,杨卫平,等.用多隐层BP网实现的CRT色度变换[J].中国图象图形学报.2000,5(6):470-473.
[10]胡涛.计算机CRT的色彩管理研究[J].计算机工程与应用,2002,(24):101-103.
[11]W.Yang,N.Liao,et al.color reproduction on CRT displays via BP neural networks under office environment[J].Journal of Beijing Institute of Technology,2003,12(4):376-380.
[12]N.Liao,Z.Gao.A comparative study of a CRT colorimetric prediction model by neural networks and the models by conventional method[J].Color Research and Application,1999,24(1):45-51.
[13]N.Tamura,N.Tsumura,et,al.masking Model for accurate colorimetric characterization of LCD[J].2003,Journal of the SID,11(2):1-7.
[14]G.Sharma.Comparative evaluation of color characterization and gamut of LCDs verus CRTs[C].2002,SPIE,4663:177-186.
[15]Y.Kwak,L.Macdonald.Characterisation of desktop LCD projector[J].Displays,2000,21(5):179-194.
[16]M.D.Fairchild,D.R.Wyble.Colorimetric characterization of the apple studio display(flat panel LCD)[R].Munsell Color Science Laboratory,1998.
[17]黄庆梅,赵达尊,等.LCD液晶显示器的颜色特性研究[J].光学技术,2005,31(2):243-245.
[18]T.Johnson.Methods for characterizing colour scanners and digital cameras.displays[J].1996,16(4):183-191.
[19]Y.Wang,H.Xu.Spectral characterization of a flat panel color scanner using PCA method[C].SPIE,2006,6033,60330J-1-5.
[20]C.Lee,E.Lee,et al.Color space conversion via gamut-based colog samples of printer[J].Journal of Imaging Science and Technology,2001,45(5):427-435.
[21]T.Johnson.Methods for characterizing colour printers[J].Displays,1996,16(4):193-202.
[22]黄庆梅,赵达尊.人工神经网络作为打印机色空间转换工具的问题[J].光学技术,2003,29(2):146-148.
[23]胡威捷,汤顺清,朱正芳.现代颜色技术原理及应用.北京:北京理工大学出版社,200731-97.
[24]王春锴.用于TFT-LCD驱动控制系统的电源管理单元的研究与设计[D].上海:复旦大学,2008.
[25]翁星光.基于LCD显示特性文件质量评测研究[D].江南大学,2007.
[26]唐剑.TFT-LCD Mura缺陷检测研究[D].北京交通大学,2008.
[27]钟云飞,陈乾坤.彩色LCD显示器色彩特性研究[J].湖南工业大学学报,2008,22(1):96-98.
[28]杨健,郝飞.显示器视觉特性化研究[J].铜仁学院学报,2008,10(5):120-122.
[29]Yasuhiro Yoshida,Yoichi Yamamoto.Color Calibration of LCDs[C].The Seventh Color Imaging Conference,2002,305-311.
[30]Nobuhiko Tamura,Norimichi Tsumura and Yoichi Miyake.Maksing Model for Accurate Colorimetric Characterization of LCD[C].IS&T/SID Tenth Color Imaging Conference,2002,312-316.
[31]Ellen A.Day,Lawrence Taplin and Roy S.Berns.Colorimetric Characterization of a Computer-Controlled Liquid Crystal Display[J].Color Res.& Appl,2004,29(5):365-373.
[32]Wen Senfar,Wu Royce.Two-primary crosstalk model for characterizing liquid crystal displays[J].Color Research and Application,2006,30(2):102-108.
[33]张显斗,徐海松.液晶显示器颜色特征化的分段分空间模型[J].光学学报,2007,27(9):1719-1724.
[34]杨卫平,廖宁放,等.基于孟塞尔系统的阴极射线管特性化新方法[J].光学学报,2004,24(8):1039-1044.
[35]Simon F T,Frost J A.A new method for the conversion of CIE colorimetric data to munsell notations[J].Color Researchand Application,1987,12(5):256-260.
[36]Tominaga S.Color notation conversion by neural networks[J].Color Research and Application,1993,18(5):253-259.
[37]Shiro Usui,Shigeki Nakauchi,et al.Reconstruction of munsell color space by a five-layer neural network[J].J.Opt.Soc.Am.(A),1992,9(4):516-520.
[38]Wyble D R,Fairchild M D.Prediction of munsell appearance scales using various color-appearance models[J].Color Research and Application,2000,25(2):132-144.
[39]T.L.V.Cheung and S.Westland,Color Camera Characterisation Using Artificial Neural Networks[R],Colour & Imaging Institute,University of Derby,Kingsway House Kingsway,Derby,United Kingdom,IS & T/SID Tenth Color Imaging Conference.
[40]Zhou.S.Q.Zhao.D.Z,BP network based color control in color printers[J],Optical technique,2000,1.
[41]Yu.H.F.Li.C.Y.Shi.J.S,Study of Color Reproduction and Color Prediction from Scanner Image to Printer Image Using Back-Propagation Neutral Networks[J],Proceedings of SPIE, 2002,4922.
[42]杨卫平,跨媒体颜色复制技术简化研究[D],北京理工大学,2005.
[43]CIE TC8-04.Chromatic adaptation under mixed illumination condition when comparing softcopy and hardcopoy images.2004.Vienna,AUSTRIA:COMMISSION INTERNATIONALE DE L'ECLAIRAGE CIE Central Bureau,ISBN 3901906xxy,http://www.cie.co.at/.
[44]ROY S.Berns.颜色技术原理[M].北京:化学工业出版社材料科学与工程出版中心,2002.
[45]束越新.颜色光学基础理论[M].济南:山东科学技术出版社,1981.
[46]大田 登著,刘中本译.色彩工学[M].西安:西安交通大学出版社,1997,173.