基于BRDF数据的金属漆颜料制品的色貌预测评估北大核心CSCD

通过双目观察与匹配心理物理学实验,对金属漆颜料制品在两种几何条件下进行了视觉评价,随后将金属漆颜料制品的视觉平均数据分别与基于双向反射分布函数(BRDF)数据和基于PR715分光辐射计测量结果的色貌模型预测结果进行最小二乘法线性拟合。实验结果表明,在不同几何条件下,BRDF数据均可以用来对金属漆颜料制品进行色貌预测;对于三属性预测而言,在明度和彩度方面,基于BRDF数据的色貌模型预测精度不如基于PR715分光辐射计数据的色貌模型预测精度,而在色调角方面,基于BRDF数据的色貌模型预测精度则高于基于PR715分光辐射计数据的色貌模型预测精度。     

基于色貌的跨媒体颜色复制

介绍了一种基于视觉匹配的跨媒体颜色复制方法。通过视觉匹配将一个环境下的一些色貌因素"映射"到另一个环境,是一种基于色貌的CRT特性化方法。该方法复制的22个Munsell色卡的平均视觉评价为6分制的5.17分。其中,红色调的复制色块视觉评价较好,蓝色调或蓝色占有较大比例的复制色块误差较大。sRGB作为目前流行的用于颜色通讯的标准色空间,在实验中也进行了比较。实验证明这种基于视觉匹配的特性化方法,已经包含了一些色貌因素,可以满足一般的应用要求,有着广泛的应用前景。     

色貌模型的人工神经网络方法的研究

色貌模型(CAM)主要解决不同观察条件、不同背景和不同环境下的颜色真实再现问题。采用人工神经网络(ANN)的方法来实现目前最新的色貌模型CIECAM02的预测,包括正向预测(从色度参数到色貌属性参数)和逆向预测(从色貌属性参数到色度参数),应用自然色系统(NCS)中的部分色样作为神经网络的训练和测试样本。由于正向输出色貌属性参数空间不是均匀的,对于网络预测精度用特殊方法评估,而对于逆向模型则可直接利用LAB色差公式评价。测试的结果表明:用神经网络对CIECAM02模型的预测达到了较高的精度。     

主观评价法在颜色研究中的应用

本文介绍了目前在色适应、色貌研究中运用方便而有效一种颜色分度方法──主观评价法。从文献中引述了其基本概念及使用原则，并列举几例说明此方法应用的普遍性和意义。     

基于LED照明的色差评价研究

截止目前,多国相关颜色科学工作者已累计提出了40多个色差公式,但是目前所有的色差公式的相关使用,都只能限定在CIE给定的观测条件。固态照明作为近些年光源研究的重点方向,相关的研究如基于LED照明条件的色差研究仍未展开。该研究遵循CIE的色差研究指导原则,首先选取CIE推荐的17种颜色中心中的五种颜色,制成实验色块样张,分别以U30,TL84和D65光源为参考光源,以LED灯泡为实验光源。选用CIE1976LUV色差公式、CIE1976LAB色差公式、CIEDE2000色差公式分别对打印色块进行理论色差值的计算。然后采用20名观察者,通过心理物理学的视觉评价实验对打印色样进行主观色差评价,最后通过标准化残差平方和(STRESS)将客观计算值与主观评价数据进行对比分析。发现实验中选用的三种色差公式在LED照明条件下的色差预测性能均不是很理想。相比而言,在LED照明条件下,CIE1976LUV公式的预测性能相对要好,CIEDE2000公式次之,CIE1976LAB公式最差。     

基于孟塞尔系统的阴极射线管特性化新方法

提出了一种新的阴极射线管特性化的方法。该方法的特点是采用“视觉匹配”方法,在反射体表面色和白发光体之间映射一些色貌因素,但没有使用任何复杂的色貌模型。是一种考虑了一些色貌因素的阴极射线管特性化方法。由于该问题的个性因素较多,采用BP神经网络实现色空间的非线性映射。实验结果表明,只要阴极射线管被标定,在办公室环境下,该方法可以改进在不同的阴极射线管上重现的颜色。采用3-7-7-3简单的网络结构;分色相样本训练。训练样本平均色差可以达到3．07L^*u^*v^*色差单位,测试样本平均色差可以达到4．55L^*u^*v^*色差单位,小于阴极射线管的最大可接受色差,结果是令人满意的。这在电子商务和办公自动化方面有广泛的用途。     

基于物体表面色的超大色差评价

基于物体表面色样本的视觉评价,测试了典型色差公式对超大色差的预测性能.由18名观察者按照倍率法对60组颜色样本对进行心理物理视觉评价实验,获得视觉色差的评价数据.利用实验数据对两个常用色差公式CIELAB和CIEDE2000以及基于色貌模型的最新色差公式CAM02-UCS和CAM02-LCD的超大色差预测性能进行了详细的比较和分析,结果表明基于色貌模型的两个公式性能优异.同时,使用本实验视觉数据考察了Attridge等人提出的幂指数模型,并得到了一致的结论.     

基于数码相机的光谱重构研究

由数码相机信号直接重构物体表面的光谱信息对不同照明条件下颜色的复现十分重要.基于普通商用数码相机分别摄取两种CIE标准光源D65和A下颜色样品的RGB信号,采用主元分析法(PCA)和多项式模型的结合算法进行物体表面色光谱反射比的重构研究.实验结果表明,同时应用两种光源下的RGB信号可有效提高光谱重构精度,并能较好地再现色样的同色异谱特性.     

基于Sigmoid-iCAM色貌模型的真实影像再现算法

为了模拟人类视觉系统成像机制,真实再现人眼对外界场景的视觉感知,为机器视觉提供符合生物视觉特性的理想图像,从色度学的角度研究了彩色图像的真实影像再现技术。针对色貌模型存在图像模糊、对比度低的缺点,在iCAM色貌模型的基础上,引入了人工神经网络的非线性Sigmoid函数,提出了基于Sigmoid-iCAM色貌模型的真实影像再现算法,并给出了该模型处理不同类型图像的控制参数统计规律。与基于CIECAM02、iCAM色貌模型的真实影像再现算法相比,该算法在对色调失真图像实现颜色复原的同时,有效地提高了图像的对比度和局部细节,很好地实现了视觉系统的颜色恒定性。     

基于模式识别技术的色貌模型评价研究

选用Munsell新标数据集,采用基于模式识别技术的圆度、色相角偏差、明度线性度与空间投影点聚集度四个指标来分别评价目前八个典型色貌模型在彩度均匀性、色相预测准确性、明度均匀性、颜色的空间再现的能力。结果表明,不同的色貌模型在四个方面的颜色再现性上表现出不同的优势。八个色貌模型对彩度预测最好的是RLAB,每个色貌模型对红色和绿色预测比较好,CAM02在色相预测整体上占优势;明度的均匀性预测都比较好,尤其是CIELAB;LLAB的颜色的空间再现能力最好。     

Color Appearance of a Patch Explained by RVSI for the Conditions of Various Colors of Room Illumination and of Various Luminance Levels of the Patch

Whenever we enter a space illuminated differently from a previous space whether in color or in illuminance, we can quickly adapt to the new atmosphere and can again perceive white for the originally white object; this is known as color constancy. This phenomenon is explained by rotation of the recognition axis of the recognized visual space of illumination (RVSI) toward the illumination color. The explanation then predicts that the color appearance of a test patch changes radically toward the opposite direction from the color of illumination when the physical property of the test patch is kept unchanged at a neutral white. This prediction was confirmed by Experiment 1, where eight different colors of illumination were employed. The test patch appeared very vivid in color and shifted toward the opposite direction from the color of the illumination. In RVSI theory the light source color mode is explained by the release of the test patch from the restriction of RVSI. The release can be achieved by increasing the luminance of the test patch and the color appearance of the patch should then return to its own color as it is no longer controlled by RVSI. In Experiment 2 these predictions were investigated by increasing the luminance of the test patch to a much higher level than that of the objects in the lit room fixed at an illuminance of about 1001x. The color appearance of the test patch indeed became the light source color and returned to the original neutral white. Emphasis was given in the course of the experiments that the subjects were observing the test patch presented in a real 3D space where the subjects also stayed inside so that they could properly construct RVSI for the space.     

Change of Color Appearance in Photopic,Mesopic and Scotopic Vision

Mesopic vision describes a range of light levels where vision is mediated by both cones and rods. The appearance of color in mesopic vision differs drastically from that in photopic vision, where only cones mediate visual information. We used a haploscopic color matching technique to investigate the color appearance under various illuminance levels, ranging from photopic to scotopic via mesopic levels. The observers did color matching between a test color chip under various illuminance levels and a matching color stimulus presented on the Cathode-Ray Tube (CRT) display under the photopic illuminance condition. The results showed that not only chroma and lightness but hue of most color chips changed with illuminance. The manner of the hue changed depended on the color of the test chip, while matching points approached a neutral gray with decrease in illuminance level for all test chips. Chroma reduced continuously with decrease of the illuminance level until 0.1 lx for reddish and yellowish color chips or until 1 lx for greenish and bluish ones. Beyond those illuminance levels, chroma was approximately constant. Lightness decreased with decreasing illuminance level for all test chips except bluish color chips, for which lightness did not decrease much in general and even increased in some cases as predicted by the Purkinje shift. The experimental results obtained in the present study provide critical features that should be considered in predicting the appearance of color at low light levels.     

Room Illuminance rather than Immediate Surrounding Luminance Determines Natural to Unnatural Color of an Object Border

The color of an object placed in a room goes through various modes of color appearance if its luminance is increased independently from the room illumination; from the natural object color to an unnatural object color, and to the light source color having two borders. We were interested in investigating the determining factor for the first border as it is useful in lighting design. It has been thought without a specific theory that it is the surrounding luminance that determines the border. According to a new concept of the recognized visual space of illumination (RVSI), the border is solely determined by the room illumination, and not by the luminance of the immediate surroundings of the object. In the first experiment of the present paper the border of a test stimulus was determined for various room illuminance while keeping the luminance of the immediate surroundings constant. It was shown that the border luminance was roughly proportional to the room illuminance, confirming the prediction based on the RVSI theory. In the second experiment the border was determined for varying luminance of the immediate surroundings while maintaining constant room illuminance. The border did not change, again confirming the prediction based on this theory.     

荧光实地印刷品Kubelka-Munk色彩预测模型

荧光油墨防伪印刷的色彩预测是防伪印刷质量控制领域十分关心的课题。本文通过分析光在荧光油墨层中的散射、吸收及荧光辐射规律,把光在印品中的传播分成紫外线和可见光两部分分别分析研究,针对多重内反射现象进行Saunderson校正,最终建立了荧光油墨实地印刷品的Kubelka-Munk模型。     

基于密度的凹版专色梯尺光谱预测模型的研究

基于密度与光谱反射率的关系建立了一种凹版专色梯尺的光谱反射率预测模型。该方法首先依据密度的定义建立实地光谱反射率与其密度的关系,并基于实地光谱反射率建立阶调光谱反射率的计算方法;然后根据密度的叠加原理,假设阶调密度与实地密度比例关系成立建立阶调密度、实地密度和承印物密度的关系;最后结合计算阶调光谱反射率的方法建立凹版专色梯尺光谱反射率预测模型;调配30种专色油墨通过凹印实打样,对预测模型通过决定系数R²和色差进行验证。实验结果表明,不同专色在同一网点面积率下实际阶调密度与实际实地密度的比例系数相同,两者的决定系数R²均大于0.98。在此关系基础上所建立的预测模型在不同网点面积率下都具有较高的决定系数,其均方根误差都小于0.01,最大色差为2.667 NBS。最后另外调配10种专色油墨在相同工艺条件下实打样样张,利用实际阶调密度与实际实地密度的比例系数,通过色差公式进而验证该模型预测专色油墨梯尺光谱反射率的精度。色差结果表明,82.12%的色差小于2.5 NBS,大部分色差在0.5～2 NBS之间,占据了总频率的58.32%,平均色差为1.58...     

基于吸光度的PET薄膜专色配方预测方法

基于吸光度与物质浓度的相关关系建立了一种PET薄膜印刷的混合油墨组分预测模型。该方法首先依据光的传播理论和光在墨层与承印材料中的多重内反射原理,利用基色和混合专色薄膜印品在黑白两色基底上的反射光谱,建立具有高透低反特性薄膜印品透射光谱的获取方法,并利用透射光谱求解吸收光谱;然后对专色吸光度与基色浓度进行回归分析,确定吸光度与基色浓度线性相关度高的特征波段,再根据郎伯比尔定律,利用特征波长处的专色样本吸光度与基色样本吸光度建立组分预测模型,求解专色样本的配方比例;最后,分析预测配方与预设浓度的偏差,并按照预测配方浓度重新打样,采用色差与光谱均方根误差对计算配方色与目标专色的匹配程度进行分析,验证模型精度。以凹印PET薄膜双基色混合样品为实验对象,对提出的方法进行测试。光谱分析表明,基色样本和专色样本光谱反射率曲线在黑白两种基底上有明显差异,但都随基色油墨浓度变化具有相同的变化规律,专色样本的光谱透射曲线随着基色浓度的变化接近浓度比例大的基色曲线,在400～580 nm区间,专色的吸收光谱随着基色油墨A浓度降低而升高,在580～700 nm区间随着基色油墨A浓度降低而降低。除570～590 nm吸收曲线相交区间以外,在可见光范围内专色油墨吸光度与基色油墨浓度之间的判定系数R2均大于0.95,平均值为0.990 0,两者具有强线性相关关系。分别选取基色A和B波长为520 nm(R2为0.994 2)和700 nm(R2为0.998 5)处的吸光度代入配色模型,并利用最小二乘法求解目标专色的配方比例。实验结果表明,六组预测浓度与预设浓度相比较,配方平均偏差为2.5%,无显著偏差。配方色样本与专色样本色差最大值为1.98,最小值为0.30,平均值为0.85,均满足GMI对专色复制的要求,其中5组专色色差小于1.5,符合专色忠实复制的要求;6组专色的光谱均方根误差最大值为2.93%,最小值为0.49%,平均值为1.40%,基本实现了同色同谱的高精度颜色复制。验证了该方法对于提高专色油墨配色精度和改善PET薄膜印刷质量具有显著效果,可为PET薄膜专色配方预测提供科学方法。     

基于色差分析的中国淡彩绘画保护性照明光源

为得到博物馆照明光源对中国传统淡彩绘画的色彩影响规律,并确定不同类型中国传统淡彩绘画的最低损害光源,以3种博物馆典型照明光源作为实验光源,分组照射中国传统淡彩绘画模型试件,对模型试件的CIE LAB色度数据进行周期性测量。基于实验数据计算不同周期色差变化值,并绘制色差随曝光量的周期性衰变曲线。进而对色差值变化数据进行回归分析,拟合得到不同光源对4种淡彩绘画颜料的相对影响函数公式,提出不同光源对各类型淡彩绘画的相对影响系数。结果表明：3种照明光源对工笔淡彩绘画的影响系数为K_金卤灯：K_卤钨灯：K_WLED=1.00：0.92：0.84;对小青绿淡彩绘画的影响系数为K_金卤灯：K_卤钨灯：K_WLED=2.05：1.71：1.65;对水墨淡彩绘画的影响系数为K_金卤灯：K_卤钨灯：K_WLED=1.17：0.94：0.91。在淡彩绘画照明保护性照明光源选择时,应选择RYGB型WLED光源。     

基于多光谱和sRGB空间的自然植物色的重现

多光谱成像技术在颜色科学方面的应用主要在于准确的记录和复制拍摄场景的颜色信息。针对目前基于光谱的颜色重现设备昂贵、复杂和实现起来困难的问题,提出了物体颜色从反射光谱空间向sRGB颜色空间变换的方案。在遵循sRGB标准的设备上重现了D65光源下的218种自然植物颜色,所显示的颜色与植物反射光谱实物对应的名称一致。对8个自然植物在9种不同光源下和在sRGB空间下的颜色进行了预测显示。所提出的方案对在网络环境下实现相同实物在不同照明环境下的场景显示提供了可行的途径。     

A Color Appearance Model Applicable in Mesopic Vision

We propose a color vision model that can be used to predict color appearance in mesopic vision as well as photopic and scotopic vision. It is based on a two-stage model which consists of the cone and opponent stages and it assumes rod intrusion at the opponent stage. The model has the following features to describe the color appearance in mesopic vision. First, it includes a gradual and nonlinear shift in spectral luminous efficiency from V (LD) to V (LD) to cope with the spectral sensitivity difference between photopic and scotopic vision and the nonlinearity of rod influence on the luminance channel. Second, the model assumes decrease of the chromatic component with decreasing illuminance to explain the reduction of saturation at low illuminance levels. Third, it assumes that red/green and yellow/blue components change with illuminance levels independently, thus explaining hue shifts with decreasing illuminance. We applied the model for color appearance simulation of natural scenes in a mesopic visual environment.