Average Color Difference Research Articles

Reflectance model for filament yarn composed of different color monofilaments

Reflected light is one of the important components that determines the color of an object. In order to predict the reflectance of the filament yarn, a model was proposed considering the reflectance, transmittance and arrangement of the monofilaments mixed in the filament. Since the fineness of a single monofilament was too small to cover the test hole of the Datacolor spectrophotometer, films with an average thickness of 0.022 mm were used to replace the monofilaments. A total of 120 samples with different layers composed of any two of four kinds of color films were designed to verify the model. The result shows that most of the color differences between the reflectances tested and calculated were below 0.8 Color Measurement Committee (2:1) units. Further, a Pearson’s correlation analysis was performed to find the significant effects on the prediction results. It was found that ΔL* and Δb* varied with the a* of the samples; Δa* varied with the L*, a* and b* of the samples. In order to improve the accuracy of the model, the correction factors were added to the reflection and transmittance of the film. The average color difference of four kinds of physical filaments was 2.48 CMC (2:1) units. The research here are of benefit to design filament product with desired color.

Color inconstancy of natural teeth measured under white light-emitting diode illuminants

ObjectiveTo evaluate color and whiteness changes of natural teeth under nine white LED illuminants recently proposed by CIE. MethodsFrom spectral reflectance factors of 36 caries-free upper central incisors, corresponding colors under illuminant D65 and CIE 1931 standard colorimetric observer were computed for all tested illuminants, using the chromatic adaptation transform CAT16. Color differences (CIELAB and CIEDE2000 units) and whiteness (WID and WIO indices) from teeth illuminated by CIE standard illuminants (D65 and A) and white LED illuminants were calculated. The appearance of 630 pairs of teeth under all illuminants was also evaluated. ResultsColor gamuts from 36 teeth under tested illuminants showed similar volume and shape in CIELAB color space, but their centers of gravity changed in many different directions with respect to colors under D65 or A illuminants: Considering colors under D65/A, color shifts produced by white LED illuminants were in the ranges 2.0–8.6/1.1–7.0 CIELAB units (1.4–8.2/0.6–6.8 CIEDE2000 units). Average WID and WIO values for the 36 teeth under different illuminants ranged from −5.8 to +19.3 and from −7.7 to +11.1, respectively. Considering 630 pairs of teeth, average color differences (mainly lightness differences) were below 0.5 CIELAB units (0.3 CIEDE2000 units), and average whiteness differences ranged from 8.1 to 10.7 for WID (23.4–25.1 for WIO). SignificanceUsing CIE LED illuminants, most changes in color and whiteness for individual teeth were above typical threshold values of perceptibility and acceptability in dentistry. However, considering pairs of teeth, the average color and whiteness differences under all tested illuminants were very similar.

Color spectra algorithm of hyperspectral wood dyeing using particle swarm optimization

To improve the accuracy and practicality of the intelligent color-matching application of wood dyeing technology, Fraxinus mandshurica veneer was selected as the dyeing material. First, based on the Friele model and Stearns–Noechel model, the model parameters were cyclically assigned to calculate the optimal fixed parameter values and predictions. Then, particle swarm algorithm was used to optimize two algorithm models, the obtained reflectance curve was fit, and the color differences were calculated according to the human eye-based CIEDE2000 color difference evaluation standard formula. Last, the two formulas to predict the color difference and spectral reflectance were compared. First, the two optimization algorithms were compared according to the size of the fitted color difference value, and then, the most accurate optimization algorithm was selected. When the model parameters were fixed, the average fitted color difference was 0.8202. After optimizing the Friele model, the average fitted color difference was 0.7287, and after optimizing the Stearns–Noechel model, the average fitted color difference was 0.6482. It was concluded that the improved Stearns–Noechel model based on particle swarm method was more accurate than the Friele model for wood color matching.

Evaluation of the influence of varnish on the color of Picasso’s Woman in Blue

Spectral and color characterizations reveal aspects of artworks which support the criteria for restoration and subsequent processes. Usually, the aim is to characterize the different compounds and their state in artworks by means of minimally invasive analyses and to provide information on deteriorated areas. It has been applied spectral analysis and other techniques to study the 2011 varnish-removal restoration of Picasso’s Woman in Blue, painted in 1901. Instrumental analyses indicated that the varnish removed in this restoration was an acrylic resin, unavailable at the time Picasso painted this artwork. This varnish acted as a filter absorbing visible light in a selective way (17–24%), modifying the color appearance of the original paint. The painting was characterized spectrally before and after the restoration, indicating that the average color difference for a set of 140 points regularly distributed throughout the painting was 3.8 Commission Internationale de l´Éclairage 1976 L*a*b* units. Specifically, on the average, the restoration substantially increased lightness, slightly decreased chroma, and caused a shift toward less yellow hues. Color changes in two different regions of the painting indicated hue-shifts in different directions. Current measurements will be useful in general conservation processes as a reference to determine future degradation.

Dominant Color Extraction with K-Means for Camera Characterization in Cultural Heritage Documentation

The camera characterization procedure has been recognized as a convenient methodology to correct color recordings in cultural heritage documentation and preservation tasks. Instead of using a whole color checker as a training sample set, in this paper, we introduce a novel framework named the Patch Adaptive Selection with K-Means (P-ASK) to extract a subset of dominant colors from a digital image and automatically identify their corresponding chips in the color chart used as characterizing colorimetric reference. We tested the methodology on a set of rock art painting images captured with a number of digital cameras. The characterization approach based on the P-ASK framework allows the reduction of the training sample size and a better color adjustment to the chromatic range of the input scene. In addition, the computing time required for model training is less than in the regular approach with all color chips, and obtained average color differences Δ E a b * lower than two CIELAB units. Furthermore, the graphic and numeric results obtained for the characterized images are encouraging and confirms that the P-ASK framework based on the K-means algorithm is suitable for automatic patch selection for camera characterization purposes.

LSSVM-based color prediction for cotton fabrics with reactive pad-dry-pad-steam dyeing

Least-square support vector machines (LSSVM) have been strongly moving from preliminary theoretical investigations into practical industrial applications. Successfully attaining a required color is a crucial task in the dyeing industry. In this paper, an LSSVM-based color quality prediction model for cotton fabrics dyed with a reactive dye using the conventional pad-dry-pad-steam (PDPS) process is proposed. The concentrations of the dye, sodium carbonate and sodium chloride, as well as the drying and steaming times were selected as the model inputs, while the CIE Lab value (L*, a*, b*) and the K/S value of the dyed fabric were used as the model outputs. The two prominent reactive dyes, Remazol Red and Remazol Navy, were used separately to dye cotton fabrics. The proposed LSSVM-based model gave mean absolute errors of less than 0.5 and 1.5 for predicting the K/S value and the CIE Lab value (L*, a*, b*), respectively. As well, the model had an acceptable average color difference ΔE* of less than 2.0 for the fabrics dyed using the PDPS process with both reactive dyes.

Evaluating whole-slide imaging viewers used in digital pathology

As a Class II medical device, whole-slide imaging (WSI) systems are emerging to succeed light microscopes used by pathologists in the past decades by digitalizing histological tissue slides into millions of pixels saved in a WSI file. Unlike the standard image file formats such as JPEG or TIFF, a WSI file usually consists of hundreds of compressed images of different magnification levels and focal planes that need to be decompressed, stitched, scaled, and colormanaged to reproduce the view demanded by the user with zooming and panning operations. Currently, most WSI files are stored in proprietary file formats, due to the lack of adopting a standard WSI file format, which hinders the development of third-party WSI viewers by making it difficult to interpret WSI files faithfully. To examine the fidelity of third-party WSI viewers, in this study, three freely available viewers, Sedeen, QuPath, and ASAP, were compared with the factory viewer, NDP, at the pixel level. A software tool was developed to register and calculate the 1976 CIE color difference for each pixel between two viewers. The average color differences were found as 1.30, 18.69, and 18.79 ΔE for Sedeen, QuPath, and ASAP, respectively.

Spectral Image Processing for Museum Lighting Using CIE LED Illuminants.

This work presents a spectral color-imaging procedure for the detailed colorimetric study of real artworks under arbitrary illuminants. The results demonstrate this approach to be a powerful tool for art and heritage professionals when deciding which illumination to use in museums, or which conservation or restoration techniques best maintain the color appearance of the original piece under any illuminant. Spectral imaging technology overcomes the limitations of common area-based point-measurement devices such as spectrophotometers, allowing a local study either pixelwise or by selected areas. To our knowledge, this is the first study available that uses the proposed CIE (Commission Internationale de l’Éclairage) light-emitting diode (LED) illuminants in the context of art and heritage science, comparing them with the three main CIE illuminants A, D50, and D65. For this, the corresponding colors under D65 have been calculated using a chromatic adaptation transform analogous to the one in CIECAM02. For the sample studied, the CIE LED illuminants with the lowest average CIEDE2000 color differences from the standard CIE illuminants are LED-V1 for A and LED-V2 for D50 and D65, with 1.23, 1.07, and 1.57 units, respectively. The work studied is a Moorish epigraphic frieze of plasterwork with a tiled skirting from the Nasrid period (12th–15th centuries) exhibited in the Museum of the Alhambra (Granada, Spain).

Deficit of luminous and normal red galaxies in cosmic voids

ABSTRACT We construct a sample of 10 680 wall galaxies and 3064 void galaxies with MR ≲ −20 by cross-referencing a void catalogue from literature with Baryon Oscillation Spectroscopic Survey (BOSS) CMASS and WiggleZ galaxies, where the CMASS survey targets redder galaxies and the WiggleZ survey targets bluer galaxies. Comparing the density profiles of the red and blue galaxies as a function of the void radius, we find that the number ratio of red-to-blue galaxies increases with distances from the void centres, suggesting a deficit of luminous and normal red galaxies in voids. We find a mean (g – r) magnitude colour of 1.298 and 1.210 for the wall and void galaxies, respectively, when considering the combined red and blue samples, which is found to be a significant difference. However, when considering the blue and red samples separately, we find no significant colour difference. We conclude that the constituents galaxies of each population, rather than intrinsic colour difference, is the main driver in the apparent average colour difference of galaxies in voids and walls, indicating a deficit of luminous and normal red galaxies in voids. Our analysis suggests that the primary environmental-dependence effect on galaxy evolution for normal and luminous galaxies between void and wall regions is manifested in the number of red galaxies, which depends on the environmental-dependent merger history. Using a semi-analytic simulation model, we can successfully reproduce the apparent colour difference between the void and wall galaxies.

Towards the design of a blending system for precoloured fibres

AbstractIn order to create a commercial system for blending precoloured fibres that will appear visually solid once combined, it is necessary to understand the maximum colour difference required between the blend components. Based on this understanding, the lowest number of primaries required to populate a given colour gamut can be determined. A series of psychophysical experiments was carried out to explore the colour difference between fibre‐blend components and whether the resulting blended samples are perceived as visually solid. Experiments were carried out with loose stock fibre, yarn and knitted samples. Generally, it was found that the likelihood a blend appeared as visually solid increased as the average colour difference between the blend components, or primaries, decreased. The value of the mean colour difference at which 50% of participants viewed the blend as being visually solid was found to be 20.8, 20.5 and 18.0 for fibre, yarn and knitted samples, respectively. Consequently, it was found that it was more difficult to obtain a solid shade with the knitted form than with the loose stock fibre form.

The optimization of color-prediction models for colored cotton fiber yarns

Color matching is necessary in the manufacture of colored fiber yarns, and its accuracy is one of the main goals in computer-aided color matching. A limited number of pre-colored fibers are blended to match the target color through predicted recipes. Of the color-prediction models applied in computer-aided color matching, two common ones, Stearns-Noechel and Friele, were selected to be modified to improve the color-prediction accuracy in this paper. The models were modified in three ways, in which the pending parameters in the models were determined through statistical analysis depending on median, wavelength, and components; thus, the Stearns-Noechel model was modified to Stearns-Noechel 1, Stearns-Noechel 2, and Stearns-Noechel 3, and the Friele model was modified to Friele 1, Friele 2, and Friele 3. The six modified models were affirmed through 261 colored fiber yarns prepared from five primary cotton fibers, including two-, three-, four-, and five-primary blends. The prediction results of 261 samples showed that Stearns-Noechel 3 had the highest accuracy among the modified Stearns-Noechel models, especially for four-primary blends with an average color difference of 0.50 Color Measurement Committee (CMC) (2:1) units, whereas Friele 1 had the highest accuracy among the modified Friele models, especially for five-primary blends with an average color difference of 0.46 CMC (2:1) units. These results indicate the modified model Stearns-Noechel 3 can be used in color prediction when there are fewer than five yarn components, and Friele 1 can be used for five components with lower color differences that better meet color-matching requirements in practical production.

K‐M theory of fabric knitted by three‐channel rotor spun wool yarn

AbstractTwo‐component and three‐component color blended yarns were spun by red, yellow, and blue wool slivers using a three‐channel rotor spun machine, and the corresponding plain fabrics were knitted. The color‐matching models of K‐M theory were built with the relative method and the least squares method, respectively. Colors and blending ratios of the fabrics were predicted by the model. The results showed that the average color differences of the samples predicted by the two methods are both about 1.0 and the mean value of the proportional error is below 3%. The least squares method has a better color‐matching effect for the three‐component sample, and the relative value method has better color‐matching results for the two‐component sample. When the tolerance range is 2.0, the pass rates of the samples predicted by either the relative value method or the least squares method reach 100%.

Determining the colorimetric attributes of multicolored materials based on a global correction and unsupervised image segmentation method.

Fast and accurate measurement of colors in multicolored prints using commercial instruments or existing computer vision systems remains a challenge due to limitations in image segmentation methods and the size and complexity of the colored patterns. To determine the colorimetric attributes (L*a*b*) of multicolored materials, an approach based on global color correction and an effective unsupervised image segmentation is presented. The colorimetric attributes of all patches in a ColorChecker chart were measured spectrophotometrically, and an image of the chart was also captured. Images were segmented using a modified Chan-Vese method, and the sRGB values of each patch were extracted and then transformed into L*a*b* values. In order to optimize the transformation process, the performance of 10 models was examined by minimizing the average color differences between measured and calculated colorimetric values. To assess the performance of the model, a set of printed samples was employed and the color differences between the predicted and measured L*a*b* values of samples were compared. The results show that the modified Chan-Vese method, with suitable settings, generates satisfactory segmentation of the printed images with mean and maximum ΔE00 values of 2.43 and 4.28 between measured and calculated values.

Flapless Postextraction Socket Implant Placement, Part 3: The Effects of Bone Grafting and Provisional Restoration on Soft Tissue Color Change-A Retrospective Pilot Study.

This article presents the results of a soft tissue color study on flapless immediate implant therapy from a sample of 23 patients who received either a provisional restoration alone or with bone grafting. The gingival color in clinical photographs was measured for the implant and for the contralateral tooth site at 2.0 and 5.0 mm below the free gingival margin using Photoshop software (Lightroom CC, Adobe). The average color difference (ΔE) values for the two groups were 2.6 and 2.4 at 2.0 mm and 1.9 and 2.5 at 5.0 mm from the free gingival margin, respectively. Approximately 80% of the sites were below the visibly perceptible threshold (ΔE = 3.1 ± 1.5) and not detectable by the human eye. The use of provisional restorations has shown positive outcomes on the stability of peri-implant soft tissue thickness and lower ΔE values. Further research is required to assess esthetic outcomes inclusive of color change relative to the clinical treatment rendered.

Color prediction of metallic coatings from measurements at common geometries in portable multiangle spectrophotometers

Illuminating and viewing geometries may strongly affect the color appearance of metallic coatings, which can be characterized accurately by bidirectional reflectance distribution function (BRDF) measurement devices. However, such devices with hundreds or even thousands of different geometries are usually expensive and complex. Accordingly, two modified models were developed in this study, based on the distribution of aluminum-flake pigments in the coatings, to, respectively, estimate the spectral radiance factors and the CIE tristimulus values of metallic coatings utilizing the measurements at 6 common geometries of portable multiangle spectrophotometers. Their performance was examined with 65 achromatic and 20 chromatic metallic coating samples under D65, A, and F11 illuminants. The average CIEDE2000 color differences over all 19 geometries were found to be less than 1.8 for both models, while the average CAM02-SCD and CIELAB color differences can, respectively, reach 1.7 and 2.0, indicating the effectiveness of our methods.

A modified single-constant Kubelka–Munk model for color prediction of pre-colored fiber blends

The goal of this work is to propose a modified single-constant Kubelka–Munk model for color prediction of pre-colored fiber blends. The original single-constant Kubelka–Munk model follows the Duncan’s additivity theorem, assuming that the optical coefficients of individual components in a turbid media are linear to their respective proportions. However, the linear assumption is invalid for the media of fiber blends due to the interactions between primary fibers, causing inaccurate color prediction of the model. Aiming at improving the accuracy, the single-constant Kubelka–Munk model was modified by employing a new additivity formula. The new additivity formula was established to achieve good linearity of the optical coefficients by modeling interactions between primary fibers as configurations. Cotton fibers blending samples were prepared to assess the color prediction accuracy. The average color difference of the proposed model was 0.91 CIEDE2000 unit, which was significantly better than that of the original model (~ 5.48). The results indicate the proposed model is much more suitable for color prediction of pre-colored fiber blends.

Comparative study on the shade matching performance of dental students under D55 and D65 light sources conditions

This study aimed to compare the effect of D55 and D65 light sources on the visual colorimetry performance of dental students by using a homemade light-source shelf. Two Vitapan 3D-Master shade guides were randomly selected. One set was used as shade guides. Ten commonly used shade tabs of 2L2.5, 2M2, 2R2.5, 3M2, 3R2.5, 3L1.5, 3R1.5, 3L2.5, 4R1.5, and 4L1.5 were selected from the other set with covered value marks and numbered from 1 to 10. After the colorimetric training, 49 undergraduate dental students were randomly divided into two groups. Each student randomly selected two of the 10 shade tabs, and the colors were subsequently matched under D65 and D55 light sources from a distance of approximately 40 cm. The average color difference (ΔE) between the color selected by each participant and the actual color of shade tabs was calculated. Paired t test was used for statistical analysis. The ΔE values between the color selected by each participant and the actual color of the shade tabs under the D55 light source varied from 0 to 6.540. The average value was 2.501. The ΔE values between the color selected by each participant and the actual color of the shade tabs under the D65 light source varied from 0 to 6.610. The average value was 2.530. No statistically significant difference was observed between the results under the two light sources (P=0.921). Both D55 and D65 daylight lamps can be used for daily dental colorimetry. These two different color temperatures showed no significant difference.

Color matching of fiber blends: Stearns‐Noechel model of digital rotor spun yarn

AbstractStearns‐Noechel model was utilized as a primary reference to study color matching principles of digital rotor spun yarn. Three primary colored (red, yellow and blue) cotton fibers were used to spin blended yarns. Spectral reflectance of the two‐component and three‐component samples was measured with data color spectrophotometer. For these samples, the Stearns‐Noechel model parameter M was determined. Four methods were employed to calculate the M value to improve accuracy of the model, 1.Classical method, named as M1; 2.Optimizing the M1 value obtained by the classical method considering the wavelength factor, named as M2; 3.Simplified M2 according to the linear correlation with the wavelength, named as M3; 4. Simplified M2 according to the segmentation correlation with the wavelength, named as M4. The study shows that average color difference of the two‐component decreases from 2.7 to 1.48, and for three‐component samples from 3.32 to 1.66, by using M2 instead of M1. While calculated using M3, the color difference of the two types of samples will be 1.73 and 2.19, correspondingly. This cannot meet color matching needs. As for M4, the average color difference of the two categories will be 1.54 and 1.91, better than the result obtained using M1 and M3, worse than M2.

Digitally reconstructing Van Gogh'sField with Irises near Arles.Part 2: Pigment concentration maps

AbstractColors in many paintings of great art historical value have changed over time, due to the combined effects of natural ageing, accumulated surface grime, and materials added during later conservation treatments. The physical restoration of the colors in such paintings is not possible. This article describes one part of work done to digitally restore the colors of Van Gogh's paintingField with Irises near Arles, dating from May 1888. We have used multispectral reflectance data to estimate absorption K and backscattering S parameters of Kubelka‐Munk 2‐constant theory. This was done for all 13 pigments known to have been used by Van Gogh in this painting, and based on this the concentration maps for each of these pigments were calculated. We validated the calculated concentration maps in several ways. For some pigments, we were able to predict spots on the painting where the pigment is expected to occur in unmixed form based on visual examination. For several other pigments, the concentration maps could be shown to agree with XRF data. Finally, for some other pigments the concentration maps were supported by additional evidence from microscopic examinations, remarks in Van Gogh's letters and from early color reproductions. For the 1.7 million pixels for which multispectral data is available, the average color difference between the calculated and measured spectral reflectance curves is CIEDE2000 = 1.05. This further confirms that the Kubelka‐Munk calculations are well suited to describe the variety of spectral reflectance on the painting.

Color prediction model for pre-colored fiber blends based on modified Stearns-Noechel function

The goal of this work is to propose a color prediction model for pre-colored fiber blends with high accuracy. The transfer function plays a vital role in an additive color mixing model. The better linearity between the transfer function and mass proportion, the higher accuracy of the model. However, the well-known Stearns-Noechel transfer function does not always hold good linearity, causing inaccurate color matching in many cases. Aiming at compensating the poor linearity, a new transfer function was established by minimizing the linear deviation. The proposed transfer function was applied to the additive model for color prediction of pre-colored fiber blends. The prediction accuracy of the proposed model was assessed by 44 samples. The average color difference was 0.63 CIEDE2000 unit, which was significantly better than the results of the Stearns-Noechel model (∼1.23) and the two-constant Kubelka-Munk model (∼1.11). These results indicate the proposed model has higher color prediction accuracy and can better satisfy the color requirement of practical production.