Heuristic analysis influence of saliency in the color diversity of natural images

The estimation of chromatic diversity of natural images is commonly quantified through the computation of the number of discernible colors and has received much attention because of the different implications it has. However, the relationship between that number and the number of colors that really attracts the attention from an observer is still not clear and has been given little attention. New concepts about salient discernible colors‐the salient chromatic diversity of images‐ and remarkable salient colors‐connected colors in the same salient image area‐are introduced as opposed to the classical number of discernible object colors, which is usually evaluated for the global image without differentiating between probable attended and non‐attended image regions. We have used different well‐known saliency models to locate the salient regions in the scenes and have heuristically studied the extent to which those models preserve the chromatic diversity of natural images. Based on a bottom‐up approach, a reduction of around 40%‐55% in the number of discernible colors were obtained, and not all saliency algorithms preserved a uniform sampling of the original color gamut. Thus, our results suggests that particularly the graph‐based visual saliency model got good low dissimilarity values in comparison with other approaches that put emphasis solely on color as the main low‐level feature. Furthermore, we have introduced a quantification scheme of the average number of remarkable salient colors appearing in the images, and have proved how the heuristic‐based analysis of salient image areas can be used to create segmented images automatically according to their salient chromatic diversity.     

Applying image‐based color palette for achieving high image quality of displays

In the highly competitive display market, manufacturers continuously develop new technologies to improve the image quality of displays. However, color measurement and visual assessment are time‐consuming to production lines. A new method to measure and improve color quality of the displays automatically therefore, is urgently needed to the manufacturers. This article proposes a familiar color correction strategy to optimize the colors of different displays by means of creating an image‐based color palette which enables color correction for familiar objects (e.g., facial skin, blue sky, or green grass) in the multidisplay systems. To produce the image‐based color palette, the 8‐bit RGB value of each pixel in an image is transformed to L*d*n* (lightness/dominant color/nondominant color) color channels, and the dominant‐color regions in an image are subsequently extracted from the dominant color (d*) channel. The memory color data of familiar objects can be set in reference monitor in advance to determine the dominant color (d*) channel. Then a series of palette colors are generated around a displayed image. The color palette will be displayed as a target for two‐dimensional colorimeter shooting to obtain the measured color data. The familiar color correction model was established based on a first‐order polynomial regression to achieve a polynomial fit between the measured color data and the reference color data on the color palette. The proposed method provides a solution to correct familiar colors on a displayed image, and maintains the original color gamut and tone characteristic in the multidisplay systems simultaneously. It is possible to achieve the preferred intent of the displayed images by using the proposed familiar color correction method. © 2012 Wiley Periodicals, Inc. Col Res Appl, 39, 154–168, 2014     

Color characteristics of Beijing's regional woody vegetation based on Natural Color System

Seasonality is a typical characteristic of Beijing's regional vegetation, and plant color is one of the most prominent visual factors of vegetation dynamic. In this research, we explored the composition and dynamic characteristics of plant color in Beijing's urban vegetation, involving the analysis of overall characteristics and respective features of leaf, flower, and fruit colors. Color data was collected from 177 woody plant species in Beijing Botanical Garden, spanning their annual life cycle, and identified with the colorimetry of the Natural Color System (NCS). Correlation and regression analyses were applied to reveal the temporal dynamic features of overall plant color richness. Cluster analysis was applied to categorize tree species based on typical colors of various plant organs. Color richness and color dispersion were introduced as two factors to measure color diversity of various tree species, applied in species evaluation by sorting and principal component analysis (PCA). Color dispersion of three‐dimensional NCS data was measured with a modified SD based on the calculation of mean spatial distance in the NCS space. Main results are as follows. The first part is plant color composition. The composition of all plant colors contains 862 NCS color species, 20 blackness species ranging from 3 to 90, 20 chromaticness species ranging from 0 to 90, 35 hue species ranging from G10Y‐B90G, and N. The second part is temporal dynamic of overall color richness. Leaf color richness and total color richness are significantly positively correlated with pentad (5‐day) sequence; flower color richness is significantly negatively correlated with pentad sequence; and fruit color richness first increases and then decreases over time. The third part is cluster analysis of tree species. Based on typical growing‐leaf color, various tree species were clustered into 6 categories; based on typical senescent‐leaf color, various tree species were clustered into 6 categories; based on typical flower color, various tree species were clustered into 15 categories; based on typical fruit color, various tree species were clustered into 7 categories. The fourth part is color diversity evaluation of various tree species with PCA. According to the PCA of flower‐leaf color diversity, the species with higher leaf color diversity and higher flower color diversity include Cotinus coggygria, Lagerstroemia indica, and Amygdalus triloba; the species with higher flower color diversity and lower leaf color diversity include Campsis radicans and Tamarix chinensis; the species with higher leaf color diversity and lower flower color diversity include Acer ginnala and Crataegus pinnatifida; the species with lower color diversity both for flower and leaf colors include Fontanesia fortune and Gleditsia sinensis. According to the PCA of leaf color diversity, the species with higher leaf color diversity in both leaf growth period and leaf senescence period include Diospyros kaki, Lagerstroemia indica and Paeonia suffruticosa; the species with higher leaf color diversity in leaf growth period and lower leaf color diversity in leaf senescence period include Amygdalus persica ‘Atropurpurea’ and Prunus virginiana ‘Canada Red’; the species with higher leaf color diversity in leaf senescent period and lower color diversity in leaf growth period include Quercus palustris, Armeniaca sibirica, and Metasequoia glyptostroboides; the species with lower leaf color diversity for the whole leaf development period include Gleditsia sinensis and Swida walteri.     

Visual comfort evaluated by number of categorical colors in a colored image

Visual comfort of a colored image of an ordinary scene was investigated in terms of the number of categorical colors contained in the image. The categorical colors were extracted by presenting the image for 5 s and asking the observer to report all the colors perceived in the image by using categorical color naming. It was found that the number of categorical colors obtained had a high negative correlation with the observer's direct estimation of visual comfort for the image. This indicates that the more colors an image contains, the less comfort for the image is felt, and suggests that visual comfort can be evaluated as a function of the number of categorical colors. The number of categorical colors was also determined colorimetrically and had a high negative correlation to the observer's direct estimation of visual comfort as well. © 2000 John Wiley & Sons, Inc. Col Res Appl, 25, 193–199, 2000     

Color preference of aged observers compared to young observers

In modern society, elderly people tend to become enthusiastic users of displays. These displays are optimized for the visual properties of young adults ignoring the specific attributes of the vision of the elderly though the existence of differences is evident. It is true, however, that most of the investigations follow a bottom‐up paradigm (changes of optical density, neural changes, etc.), and their results are too ambiguous to be taken into account directly in display optimization for the elderly, partly because there is a long‐term adaptation, which apparently recompensates for some of the changes in the human visual system. Preference is a high‐level psychological factor having a very important impact on the acceptability of color displays. The present article follows a top–down methodology to investigate these age‐related differences directly. The first part concentrates on functionalities of the human visual system: white point preference, chroma perception, unique, and preferred hues. Results confirm the long‐term compensation theory for the elderly observers along with the chromatic content decrease of perceived colors. The second part investigates preference differences in case of photorealistic images in terms of global and local contrast, white point, average chroma, and the effect of several image color manipulation techniques. Results indicate significant differences between young and aged observers' color image preference, some of which can be explained with neuro‐physiological changes, others may be attributed to cultural implications. © 2008 Wiley Periodicals, Inc. Col Res Appl, 33, 381–394, 2008     

Near perfect visual compensation for atmospheric color distortions

Light is scattered and absorbed in the atmosphere producing visual effects that increase with viewing distance. Contrast and chromatic diversity decrease with distance, thereby modifying the way objects and scenes are perceived. Although some perceptual compensation to produce color constancy of individual surfaces has been reported, it is unclear to what extent unknown original images can be visually inferred from their distorted versions, that is, how much these effects can be discounted by the visual system. We investigated this issue with a paradigm akin to the paper-matching paradigm used in color constancy studies but with complex natural images. Hyperspectral data from 11 natural scenes were used to simulate their colors for distances up to 2000 m using a precise physical model of the effects of atmosphere. In each trial of the experiment, observers viewed the range of simulated images derived from one scene displayed on a calibrated monitor and selected the one perceived as more natural, without color distortion. Pooling data across scenes and observers showed that the image selected as more natural was very close to the undistorted one, corresponding to a simulated distance of only 2 m. These results suggest that observers are sensitive to changes in the naturalness of colors due to atmospheric effects and, crucially, can retrieve the original chromatic content with good accuracy revealing an efficient form of color constancy.     

Research on color harmony of building façades

In this study, the 28 primary colors and 11 complementary colors suggested by Chang et al in their investigation on building colors in Wanhua District of Taipei City were taken as color samples. The two-color combination mode was adopted to obtain 308 simulation photos, and two-color harmony was discussed from the perspective of visual evaluation using psychophysical tests. This study explored building façade color harmony in the CIELAB color space, and the relationship between the color attributes (hue, lightness, and chroma) and the color harmony, and between the differences of the color attributes and color harmony. It found that a high lightness of a building's primary color is associated with a high level of building color harmony, while the color harmony is reduced when the color falls in the green or blue sector in the CIELAB color space; a greater lightness difference between building façade colors is associated with a higher level of building color harmony, while the colors are disharmonized when they tend to the blue sector in the CIELAB color space. The contribution of this study is to summarize the principles for the application of building color harmony in urban renewal, and proposed suggestions on building color harmony in the urban renewal process.     

Color and visual complexity in abstract images: Part II

There are a limited number of studies examining color, visual complexity, and visual interest together, and one of the recent studies that tried to bring a new understanding to the association between color, visual complexity, and visual interest was the first part of the current study. Most of the well‐known color studies researching the effects of color on psychology, physiology, emotion, mood, attention, well‐being, visual complexity, and visual interest used isolated color patches that might be lacking in reflecting the dominating factors. Thus, the aim of this study was to find the relationship between visual complexity, visual interest, and color difference (ΔE) values of colors in abstract images, and it was hypothesized that, as the average ΔE value of colors in an abstract image increases, visual interest and visual complexity will increase until reaching a threshold where visual interest and visual complexity start to decrease. In order to test the hypothesis, a new abstract image was generated and colored. The generated abstract image was rated by 120 undergraduate students from the Faculty of Art, Design and Architecture. As the results of the study indicated, there was an inverted U‐curve relationship between average ΔE values and both visual interest and visual complexity in abstract images.     

Hierarchical emotional color theme extraction

Artists usually carefully select different colors in artistic work so as to convey special visual and emotional feelings. Color theme extraction techniques can help users to acquire the color styles in an image. However, current color theme extraction methods ignore the emotional factors, and they can only provide a single theme result for an image as well, neither of which meet people's favor on different colors under different mood states. This article introduces the conception of emotional color theme, introduces the color emotion theory into color theme extraction, and proposes a novel emotion color theme extraction framework. To achieve these goals, we perform the theme extraction with emotion value of each pixel instead of color value. The emotional discrepancy is proposed between the colors in the theme to evaluate a color theme quality and prove it with color theme dataset. Then a data driven approach is adopted to optimize the color theme. Different from previous optimizing strategies, we built the emotional relation between the target theme and the candidate theme. Our method can enhance the color editing results of previous methods, e.g., for color transfer our results can demonstrate hierarchical emotions from a single reference image. © 2015 Wiley Periodicals, Inc. Col Res Appl, 41, 513–522, 2016     

Color naming experiments using 2D and 3D rendered samples

This article describes a color naming experiment using 2D and 3D rendered color samples. Conventional color naming experiments using a priori clues generally involve 2D clues such as color patches. However, in real‐world scenes, most objects have 3D shapes whose colors are affected by illumination effects such as shadows and gloss. We use 2D and 3D rendered samples as clues in the experiments, and analyze the relationship between color terms and object surfaces. First, we develop a color term collection system that can produce 218 test colors. We render the color images of a flat disk as a 2D sample and a sphere as a 3D sample on a calibrated display device. It is supposed that the 2D and 3D surfaces with the same object color are obtained under the same conditions of viewing and illumination. The results of color naming experiments show that there are differences for color terms between 2D and 3D samples. Important findings are as follows: (1) brighter color terms tend to be chosen for the 3D samples than the 2D samples, when observing achromatic colors, (2) achromatic color terms are chosen for 3D samples having low saturation, and (3) for chromatic colors, a darker color term is generally chosen in comparison to the corresponding 2D samples of the same color. These properties become more prominent by changing the illumination angle from 0° to 45° to the surface normal. © 2014 Wiley Periodicals, Inc. Col Res Appl, 40, 270–280, 2015     

A planar-vector based color space for graphic arts color analysis and reproduction

A new color space is developed, based on a model of a trigonally symmetric, planar arrangement of three response vectors, which first separates a visual stimulus via tristimulus values into achromatic and chromatic components and then recombines them. Because three visual response vectors are mutually opposed, the model is called the mutually opposed, trichromatic response (MOTR) model. The MOTR model does not introduce any new visual concepts but rather uses established concepts in an empirically different way. The MOTR model directly represents a color as a combination of two chromatic primaries and one achromatic primary. It is, therefore, particularly useful for halftone color reproduction based on two primary colors plus black, sometimes referred to as 100% GCR (gray component replacement). It is useful also because of its linearity properties between lightness and the chromatic and achromatic contents. The MOTR model's color space is applicable to existing color-order systems and provides a graphic representation often better than CIELAB and CIELUV do. The MOTR model is also consistent with many known visual characteristics such as unique hues, illuminant adaptation, wavelength discrimination, and defective color vision. The model as described herein is limited to the comparison of related colors under a given photopic luminance.     

Optimal employment of color attributes to achieve saliency in icon matrix designs

With the wide use of smart devices, through which information is presented in vast quantities, objective guidelines are needed to enable designers to choose appropriate colors for information display. The purpose of this study is to determine which colors are the most eye‐catching in displays that employ icon matrices and thereby provide empirical grounds for strengthening the visual information structure of interface designs. Three attributes of color, which include hue, tone, and color combinations, are examined to optimize the color saliency in information displays. An eye‐tracking study was conducted to evaluate saliency objectively by analyzing fixations of visual attention. Based on the hue‐saturation‐brightness color system, a 5‐by‐5 matrix of 25 color patches was adopted to generate 21 color stimuli. Part I of the study focused on hue and indicated that warm colors are perceptually more eye‐catching relative to cool and neutral colors. Part II of the study investigated tonal influences and revealed that highly saturated colors provoked the greatest visual magnetism against a black background across all hue groups, although there was an alternative tendency for a blue hue. Contrary to expectations, no distinctive patterns were observed among brightness groups. With regard to color combination, Part III of the study provided empirical verification that high contrast between a foreground and a background generates a more dominant conspicuity. The results of the present study can be applied in designing electronic interfaces that display icon matrices to create effective communication by guiding visual attention and increasing aesthetic satisfaction. © 2014 Wiley Periodicals, Inc. Col Res Appl, 40, 429–436, 2015     

A new color image space HRU related to the CIEL*a*b* color space

The color image space with one kind of merit color image scales (WIP) is derived using the psychophysical method of magnitude estimation method, usually used in visual assessment of color appearance, which can be used to describe the color image meanings of colors of works of art in parallel with those of people. The results show that a new color image space HRU is developed, and the relativity between this space and the CIEL*a*b* color space is also discussed. And, a good relationship between the HRU color image space and the CIEL*a*b* color space can be found. This may be of great advantage to the new color image space HRU in predicting the color image for single color based on the color space CIEL*a*b*. © 2009 Wiley Periodicals, Inc. Col Res Appl, 34, 452–457, 2009     

Color reproduction of authenticable luminescent backlit transmissive color images

We propose a method for creating authenticable color images under UV excitation backlight by printing an invisible luminescent white emissive ink layer on the verso side of a transmissive substrate and a classical cmy image on the recto side of the substrate. In order to obtain a backlit image whose colors are as close as possible to the original image colors, we map the input image sRGB color gamut into the gamut formed by the emissive white source attenuated by the classical ink halftones. The relationship between surface coverages of the classical cmy ink halftones and the resulting backlit colors is obtained by accounting for the transmission of the emissive white through the paper substrate and through the classical cmy ink halftones. The transmittance of the classical ink halftones is modeled by a new halftone absorbance prediction model. The lightness range of the luminescent backlit color gamut is expanded by printing a black and white UV‐absorbing instance of the original color image in superposition with the luminescent white emissive layer and in registration with the cmy image printed on the recto side. Luminescent backlit color images provide a high anticounterfeiting security, as they combine a verso printed invisible luminescent ink layer, a verso printed black and white instance of the original image and a matching recto printed cmy instance of the original image. The resulting luminescent backlit image colors are close to the original image colors only when observed in transmission mode under UV light. © 2013 Wiley Periodicals, Inc. Col Res Appl, 39, 331–340, 2014     

A novel method for fabric color transfer

Fabric color design is a complex process in textiles and clothing industry. A new method for fabric color selection and transferring is proposed in this study. An automatic way to select the colors from the natural images is developed for fabric color design. Based on these colors, a fabric image is then used for color transferring. The fabric image is processed by a bias field estimation operation, and the membership function of the color deviations of the image has been obtained. According to the selected colors and the color membership function, the fabric image colors can be changed and transferred to a new image that preserves the similar texture appearance but with significantly different color effects. The experimental results confirm the effectiveness of the proposed method. © 2014 Wiley Periodicals, Inc. Col Res Appl, 40, 304–310, 2015     

Color constancy from invariant wavelength ratios. III: Chromatic adaptation theory,model and tests

A theory of chromatic adaptation is derived from Parts I and II, and presented in terms of relative wavelength, purity, and radiant power, leading directly to a predictive model of corresponding hue, chroma, and lightness. Considering that even simple animals have effective color vision and color constancy, the aim was to develop a simple model of complete adaptation. The model is tested against well‐known data sets for corresponding colors in illuminants D65, D50, and A, and for small and large visual fields, and performs comparably to CIECAM02. Constant hue is predicted from Part I's mechanism of color constancy from invariant wavelength ratios, where constant hues shift wavelength linearly with reciprocal illuminant color temperature. Constant chroma is predicted from constant colorimetric purity. Constant lightness is predicted from chromatic adaptation of spectral sensitivity represented by power ratios of complementary colors (rather than cone responses which lack spectral sharpening). This model is the first of its type and is not formatted for ease of computation. © 2010 Wiley Periodicals, Inc. Col Res Appl, 2010     

A comparative analysis of the characteristics and images of costume colors in the traditional plays of Korea,China, and Japan

The purpose of this study was to analyze quantitatively the characteristics and images of costume colors in the traditional plays of Korea, China, and Japan. The study focused on the Korean Masque, Beijing Opera, and Kabuki costume colors based on a selection of 1135 color samples. The collected source data were selected by extracting digital color data by using the Eyedropper Tool of Photoshop 7.0. The RGB color data were transformed to H V/C and the attributes of hue and tone were analyzed. Color images were analyzed with the color image scales of IRI Color Design Institute and Shigenobu Kobayashi to increase the validity of the evaluated images. As a result, the “five element colors (red, yellow, purple–blue (PB), white, and black)” from the theory of “Yin‐Yang Wu‐Xing” were used in the common stage costume colors of the Korean Masque, Beijing Opera, and Kabuki. Red, a preferred Asian color, was used most frequently in the costumes of these three traditional plays. A comparison of the traditional stage costume colors in the three northeast Asian countries revealed a difference in tones rather than in hues. First, the Korean Masque frequently used white in accordance with the tradition of white‐clad people and the cultural view of colors in which natural colors were preferred. Additionally, in the Masque, Koreans used colors based on the theory of Yin‐Yang Wu‐Xing with high‐chroma tones. On the other hand, the Beijing Opera exhibited the gorgeous and strong color images of China, by adopting high‐chroma colors in the Five Element Color: R, Y, PB, white, and black. Last, in the Kabuki costumes, a variety of white, black, dull, light, dark, strong, vivid, deep, bright, and grayish tones played an important role in showing various color images. The costume color images of the traditional plays of the three countries revealed that all shared the use of dynamic, springy/casual, and gorgeous images in the strong contrast of five element colors. Regarding the differences, the Korean Masque exhibited natural images in favor of natural colors, whereas the Kabuki displayed modern, decent/formal images by using dull, dark, and grayish colors. The study results suggested that the three countries commonly used five element colors from the theory of Yin‐Yang Wu‐Xing, but that their color images differed in terms of the tones used and the techniques for color combination. These results reflect that colors in the traditional costumes of the three countries are affected by their cultural codes, thereby representing the characteristics ofcertain peoples and cultural circles. © 2011 Wiley Periodicals, Inc. Col Res Appl, 2011     

Skin‐color correction method based on hue template mapping for wide color gamut liquid crystal display devices

The development of wide color gamut (WCG) liquid crystal display (LCD) plays an important role in the high‐quality television (TV) field. Nowadays, people want their TV or display devices to have the capability of showing vivid colors while keeping skin colors as natural as they remember. Therefore, it is necessary to develop color‐correction technologies for WCG LCD system. A new color‐correction method named “natural skin‐color mapping algorithm” (NSCMA) for WCG LCD is proposed in this study. It can solve the skin‐color contour problem in color‐corrected images with simple skin‐color detection. Its development is based on the concepts of performing color mapping between source hue colors and target hue colors on each hue page. The polynomial regression is also applied to calculate the color mapping conversion matrices. Two color mapping factors called template‐size factor and tone‐compression factor are designed in NSCMA. The template‐size factor is used to adjust target template sizes adequately. The tone‐compression factor is designed to control the degrees of image enhancement. For facial skin‐color pictures, the appropriate settings of template‐size factor and tone‐compression factor will get suitable color image rendering on the WCG LCD. It is demonstrated that the WCG LCD can be corrected to show vivid color pictures and keep facial skin colors as natural as possible when the proposed NSCMA is performed. © 2010 Wiley Periodicals, Inc. Col Res Appl, 2011