Color Constancy Research Articles

Construction of durable and original color constancy photochromic cotton fabrics by a facile esterification strategy

Cotton-based photochromic textiles have emerged as a representative smart textile, yet constrained by the complicated process, deep original color and limited durability in practical applications. Here, spiropyran (SP) with a simple carboxy linking group was directly covalently immobilized onto cotton fabric through a facile and effective esterification strategy. Benefit from the simple structure, this modification process does not affect the basic properties and color of the fabric. In addition, the SP-cotton fabrics exhibited excellent reversible photochromic properties from colorless to violet and allowed to print exquisite patterns with long-distance control by switching light resource and photomasks. Moreover, the stable ester covalent bond endowed these SP-cotton fabrics with appropriate washing fastness and photo fatigue resistance. This study not only added value to cotton, but also paved a rational and general approach for the construction of durable and original color constancy photochromic textiles, providing a potential candidate for brand protection, anti-counterfeiting, and fashion textiles. • A novel photochromic fabric was prepared by directly covalently immobilizing spiropyran (SP) contained carboxy linking group. • A simple and extensible immobilized strategy was proposed for constructing original color constancy photochromic textiles. • SP cotton with photochromic performance from colorless to violet and allowed to print patterns with long-distance control.

Brain-like retinex: A biologically plausible retinex algorithm for low light image enhancement

Retinex theory was first proposed by Land and McCann [1], where retinex is a portmanteau derived from the words of retina and cortex, implying that both the retina and cerebral cortex may participate in the perception of lightness and color. However, there are no recent reports on how the retina and visual cortex perform retinex decomposition. In this paper, we propose a biologically plausible solution to retinex decomposition. We develop an algorithm motivated by the primate’s retinal circuit to detect textural gradients, design an algorithm originating from the visual cortex to extract image contours, and thus split image edges into image contours and textural gradients. Then, we establish a variational model for retinex decomposition by using image contours and textural gradients to encode discontinuities in illumination and variations in reflectance, respectively. We also apply the proposed retinex model to low light image enhancement, high dynamic resolution image toning, and color constancy. Experiments show consistent superiority of the proposed algorithm. The code is available at Github.

Dive into Illuminant Estimation from a Pure Color View

Illuminant estimation is critically important in computational color constancy, which has attracted great attentions and motivated the development of various statistical- and learning-based methods. Past studies, however, seldom investigated the performance of the methods on pure color images (i.e., an image that is dominated by a single pure color), which are actually very common in daily life. In this paper, we develop a lightweight feature-based Deep Neural Network (DNN) model—Pure Color Constancy (PCC). The model uses four color features (i.e., chromaticity of the maximal, mean, the brightest, and darkest pixels) as the inputs and only contains less than 0.5k parameters. It only takes 0.25ms for processing an image and has good cross-sensor performance. The angular errors on three standard datasets are generally comparable to the state-of-the-art methods. More importantly, the model results in significantly smaller angular errors on the pure color images in PolyU Pure Color dataset, which was recently collected by us.

Computational modeling of color perception with biologically plausible spiking neural networks.

Biologically plausible computational modeling of visual perception has the potential to link high-level visual experiences to their underlying neurons' spiking dynamic. In this work, we propose a neuromorphic (brain-inspired) Spiking Neural Network (SNN)-driven model for the reconstruction of colorful images from retinal inputs. We compared our results to experimentally obtained V1 neuronal activity maps in a macaque monkey using voltage-sensitive dye imaging and used the model to demonstrate and critically explore color constancy, color assimilation, and ambiguous color perception. Our parametric implementation allows critical evaluation of visual phenomena in a single biologically plausible computational framework. It uses a parametrized combination of high and low pass image filtering and SNN-based filling-in Poisson processes to provide adequate color image perception while accounting for differences in individual perception.

Evolving convolution neural network by optimal regularization random vector functional link for computational color constancy

To automatically eliminate the influence of external illumination on the acquired image, restore the true color information of the object, and provide true and stable color features for computer vision tasks, our work is based on the convolutional neural network (CNN) VGG19 for feature extraction of image information. Moreover, it proposes a model based on atom search optimization improved by chaotic-logistic maps to optimize the regularization random vector functional link (RRVFL). The optimized RRVFL replaces CNN’s regression layer to estimate the image illumination and then restore the image. First, we used chaotic-logistic maps to optimize the initial value of search agent of the atom search optimization (CASO) algorithm and then used the search agent segment mapping method proposed in our study to simultaneously optimize the number of nodes in hidden layer, regularization factors, input weight, and the bias of RRVFL hidden layer. This avoids the problem of serious fluctuation of prediction results and low accuracy caused by the randomness of the RRVFL parameters. After the predicted illumination information was obtained with the optimized RRVFL, the image was restored using the diagonal transformation method. Comparative experiments showed that the average angular errors of the illumination estimation of the V19-CASO-RRVFL algorithm proposed in our study were 1.73, 0.30, 0.3406, and 0.0976 lower than those of the gray-edge-2, GE-CASO-RRVFL, DS-Net, and color constancy on deep residual learning algorithms, respectively.

Using convolutional neural network models illumination estimation according to light colors

One of the important problems in digital images is that the object's color changes as the light source's color changes. Color constancy methods are used to solve these problems. Color constancy is the ability to detect object color in an image under lighting sources accurately. The light color of the image is estimated by calculating color constancy. Many statistical and learning-based approaches related to calculating color constancy have been presented. In recent years, deep learning algorithms, one of the learning-based approaches, have been used to calculate color constancy. This study aims to estimate the illumination of images obtained in varying light colors with deep learning methods. This study was applied to the agricultural sector. The illumination estimation was performed using the 3-fold cross-validation method with VGG16, EfficientNet-B0, ResNet50, MobileNet, DenseNet121, and GoogLeNet models. A transfer learning approach was adopted in this study. Illumination estimation was applied to a new data set. The median value of the angular error (AE) metric performed well in all experimental results. The lowest AE values were obtained in the proposed GoogLeNet model. This model AE values: the mean was 2.220, the median was 2.126, the trimean was 2.006, and the maximum was 6.596 degrees. In addition, the number of images with AEs below 3.0˚ constituted 77.13% of all images. The results of the Friedman and Wilcoxon signed rank tests confirmed the effectiveness of the proposed GoogLeNet model in illumination estimation.

Strong green-emission phosphor Ca2SiO4:Eu2+ to achieve the color constancy and high luminous efficiency for white LEDs

The study uses the green phosphor of Ca2SiO4:Eu2+ to achieve the color constancy for the dual-film remote phosphor white-LED model at low as well as big CCTs. The utilized green phosphor Ca2SiO4:Eu2+ was prepared using the water-soluble-silicon liquid phase precursor method with the doped Eu2+ ion concentration of 3 mol%. The Ca2SiO4:Eu2+ phosphor emits strong green light with emission intensity focused at 502 nm wavelength, and a wide stimulation band of colors of 225 nm – 450 nm. After applying the Ca2SiO4:Eu2+ green phosphor and modifying its concentration, the modified color and luminous performances can be observed. The better color uniformity and higher luminescence efficiency can be obtained by increasing the percentage of Ca2SiO4:Eu2+ in the phosphor configuration. Meanwhile, the color rendering metrics tend to reduce slightly when the concentration of Ca2SiO4:Eu2+ is over 10% wt.

Multi-scale fusion framework via retinex and transmittance optimization for underwater image enhancement

Low contrast, poor color saturation, and turbidity are common phenomena of underwater sensing scene images obtained in highly turbid oceans. To address these problems, we propose an underwater image enhancement method by combining Retinex and transmittance optimized multi-scale fusion framework. Firstly, the grayscale of R, G, and B channels are quantized to enhance the image contrast. Secondly, we utilize the Retinex color constancy to eliminate the negative effects of scene illumination and color distortion. Next, a dual transmittance underwater imaging model is built to estimate the background light, backscattering, and direct component transmittance, resulting in defogged images through an inverse solution. Finally, the three input images and corresponding weight maps are fused in a multi-scale framework to achieve high-quality, sharpened results. According to the experimental results and image quality evaluation index, the method combined multiple advantageous algorithms and improved the visual effect of images efficiently.

Understanding insect colour constancy.

Colour constancy is the ability to recognize the colour of objects despite spectral changes in the natural illumination. As such, this phenomenon is important for most organisms with good colour vision, and it has been intensively studied in humans and primates. Colour constancy is also documented for several species of insects, which is not surprising given the ecological importance of colour vision. But how do insects, with their small brains, solve the complex problem of colour vision and colour constancy? In an interspecies approach, this review reports on behavioural studies on colour constancy in bees, butterflies, moths and humans, corresponding computational models and possible neurophysiological correlates. This article is part of the theme issue 'Understanding colour vision: molecular, physiological, neuronal and behavioural studies in arthropods'.

Colour vision in nocturnal insects.

The ability to see colour at night is known only from a handful of animals. First discovered in the elephant hawk moth Deilephila elpenor, nocturnal colour vision is now known from two other species of hawk moths, a single species of carpenter bee, a nocturnal gecko and two species of anurans. The reason for this rarity—particularly in vertebrates—is the immense challenge of achieving a sufficient visual signal-to-noise ratio to support colour discrimination in dim light. Although no less challenging for nocturnal insects, unique optical and neural adaptations permit reliable colour vision and colour constancy even in starlight. Using the well-studied Deilephila elpenor, we describe the visual light environment at night, the visual challenges that this environment imposes and the adaptations that have evolved to overcome them. We also explain the advantages of colour vision for nocturnal insects and its usefulness in discriminating night-opening flowers. Colour vision is probably widespread in nocturnal insects, particularly pollinators, where it is likely crucial for nocturnal pollination. This relatively poorly understood but vital ecosystem service is threatened from increasingly abundant and spectrally abnormal sources of anthropogenic light pollution, which can disrupt colour vision and thus the discrimination and pollination of flowers.This article is part of the theme issue ‘Understanding colour vision: molecular, physiological, neuronal and behavioural studies in arthropods’.

NPR 이미지의 색 항상성 - 애니메이션 장면을 중심으로

Animation is a video made up of non-photorealistic expressive still images considering the human perspective. There is an increasing need to clarify the color expression characteristics of NPR images to facilitate the animation production. This study paid attention to the color constancy, which recognizes the color of the surface as the same color even when the lighting color of the environment changes to various colors. Therefore, the color constancy was researched in NPR animation images. The subjects produced animation images including the character and the background, considering the color composition including 6 objects (apple, banana, leaf, soil, tree, skin) based on the basic color term. According to the change of time, the colors were set to 12:00 PM (6500K) at noon, 6:00 PM (4000K) at sunset, and 12:00 PM (20000K) at night. As a result, the CIE LAB values of the object color shown in the NPR animation image were obtained, and the Color Constancy Index (CI) was calculated according to the color temperature change to confirm the effect of color constancy. As a result, the CI of all object colors except for the sky at 4000K was higher than 0.6, confirming the effect of the color constancy. In addition, at 20000K, lower than CI=0.5 was obtained for all objects except for the skin, confirming the action of the low color constancy. This study is different from the existing color constancy studies focusing on photographed images, and it can help to build a color database required from the viewpoint of animation creators.

Fusion-based underwater image enhancement with category-specific color correction and dehazing.

Underwater imaging is usually affected by water scattering and absorption, resulting in image blur and color distortion. In order to achieve color correction and dehazing for different underwater scenes, in this paper we report a fusion-based underwater image enhancement technique. First, statistics of the hue channel of underwater images are used to divide the underwater images into two categories: color-distorted images and non-distorted images. Then, category-specific combinations of color compensation and color constancy algorithms are used to remove the color shift. Second, a ground-dehazing algorithm using haze-line prior is employed to remove the haze in the underwater image. Finally, a channel-wise fusion method based on the CIE L* a* b* color space is used to fuse the color-corrected image and dehazed image. For experimental validation, we built a setup to acquire underwater images. The experimental results validate that the category-specific color correction strategy is robust to different categories of underwater images and the fusion strategy simultaneously removes haze and corrects color casts. The quantitative metrics on the UIEBD and EUVP datasets validate its state-of-the-art performance.

Angle-Retaining Chromaticity and Color Space: Invariants and Properties.

The angle-retaining color space (ARC) and the corresponding chromaticity diagram encode information following a cylindrical color model. Their main property is that angular distances in RGB are mapped into Euclidean distances in the ARC chromatic components, making the color space suitable for data representation in the domain of color constancy. In this paper, we present an in-depth analysis of various properties of ARC: we document the variations in the numerical precisions of two alternative formulations of the ARC-to-RGB transformation and characterize how various perturbations in RGB impact the ARC representation. This was done empirically for the ARC diagram in a direct comparison against other commonly used chromaticity diagrams, and analytically for the ARC space with respect to its three components. We conclude by describing the color space in terms of perceptual uniformity, suggesting the need for new perceptual color metrics.

Developmental changes in colour constancy in a naturalistic object selection task.

When the illumination falling on a surface change, so does the reflected light. Despite this, adult observers are good at perceiving surfaces as relatively unchanging-an ability termed colour constancy. Very few studies have investigated colour constancy in infants, and even fewer in children. Here we asked whether there is a difference in colour constancy between children and adults; what the developmental trajectory is between six and 11 years; and whether the pattern of constancy across illuminations and reflectances differs between adults and children. To this end, we developed a novel, child-friendly computer-based object selection task. In this, observers saw a dragon's favourite sweet under a neutral illumination and picked the matching sweet from an array of eight seen under a different illumination (blue, yellow, red, or green). This set contained a reflectance match (colour constant; perfect performance) and a tristimulus match (colour inconstant). We ran two experiments, with two-dimensional scenes in one and three-dimensional renderings in the other. Twenty-six adults and 33 children took part in the first experiment; 26 adults and 40 children took part in the second. Children performed better than adults on this task, and their performance decreased with age in both experiments. We found differences across illuminations and sweets, but a similar pattern across both age groups. This unexpected finding might reflect a real decrease in colour constancy from childhood to adulthood, explained by developmental changes in the perceptual and cognitive mechanisms underpinning colour constancy, or differences in task strategies between children and adults. HIGHLIGHTS: Six- to 11-year-old children demonstrated better performance than adults on a colour constancy object selection task. Performance decreased with age over childhood. These findings may indicate development of cognitive strategies used to overcome automatic colour constancy mechanisms.

The Development of Color Perception and Cognition.

Color is a pervasive feature of our psychological experience, having a role in many aspects of human mind and behavior such as basic vision, scene perception, object recognition, aesthetics, and communication. Understanding how humans encode, perceive, talk about, and use color has been a major interdisciplinary effort. Here, we present the current state of knowledge on how color perception and cognition develop. We cover the development of various aspects of the psychological experience of color, ranging from low-level color vision to perceptual mechanisms such as color constancy to phenomena such as color naming and color preference. We also identify neurodiversity in the development of color perception and cognition and implications for clinical and educational contexts. We discuss the theoretical implications of the research for understanding mature color perception and cognition, for identifying the principles of perceptual and cognitive development, and for fostering a broader debate in the psychological sciences.

Evaluation of the Impact of Different Disinfectants on Color Stability of Denture Base Materials: A Comparative Study

Aim of the current research was to assess the effect of different disinfectant solutions on the color steadiness of denture base materials. Ninety samples were made employing the Meliodent and ProBase Hot heat cure denture base resin materials. Metallic disks measuring 3 mm thick as well as 20 mm in diameter were utilized to make the samples. Forty-five samples of every denture base material were immersed in three chemical disinfectants: 2% alkaline glutaraldehyde, 0.5% chlorhexidine gluconate, 0.5% sodium hypochlorite solution as well as distilled water that served as a control. Color scoring was performed at baseline and following first, 14th as well as 21 days. Color was calculated as per CIE L*a*b* color scale as related to standard source C alongside a white background on a reflection spectrophotometer. Subsequent to the immersion of Meliodent heat-cured denture base resin materials within distilled water, the baseline color steadiness score was 6.16 ± 0.12, 6.12 ± 0.08 on the first day, 5.98 ± 0.06 on day 14 and 5.74 ± 0.14 on day 21. Within 2% alkaline glutaraldehyde the baseline color constancy mean value was 6.12 ± 0.28, 6.08 ± 0.04 on the first day, 5.04 ± 0.11 on day 14 and 4.22 ± 0.14 on day 21. Within 0.5% chlorhexidine gluconate the baseline color constancy mean value was 6.14 ± 0.09, 6.02 ± 0.12 on the first day, 5.38 ± 0.19 at 14th day and 4.84 ± 0.10 on day 21. In 0.5% sodium hypochlorite solution the baseline color steadiness mean value was 6.16 ± 0.18, 6.10 ± 0.06 on the first day, 4.98 ± 0.26 on 14th day and 3.14 ± 0.14 on day 21. ProBase Hot heat cure denture base resin materials after immersion in Distilled water delineated the baseline color constancy mean value was 6.48 ± 0.02, 6.34 ± 0.09 on the first day, 6.08 ± 0.14 on 14th day and 5.88 ± 0.07 on day 21. Within 2% alkaline glutaraldehyde the baseline color constancy mean value was 6.42 ± 0.16, 6.18 ± 0.11 on the first day, 5.12 ± 0.18 at 14th day and 4.24 ± 0.24 on day 21. Within 0.5% Chlorhexidine gluconate the baseline color steadiness mean value was 6.48 ± 0.10, 6.26 ± 0.08 on the first day, 5.48 ± 0.11 on 14th day and 4.88 ± 0.06 on day 21. In 0.5% sodium hypochlorite solution the baseline color steadiness mean value was 6.44 ± 0.12, 6.14 ± 0.08 on the first day, 4.98 ± 0.16 on 14th day, and 3.34 ± 0.06 on day 21. There was a statistically noteworthy difference amid the dissimilar time points in each of the three disinfectants. In the confines of the inherent limitations of this research, it can be inferred that the color steadiness of either denture base resins was enhanced following immersion in 0.5% chlorhexidine gluconate in pursuit by 2% alkaline glutaraldehyde as well as 0.5% sodium hypochlorite solution. Distilled water exerted the smallest amount of influence on the color change of the samples of denture base resins. Techniques that help infection prevention influence denture disinfection. Currently, there is a dearth of commercially marketed agents that are specifically intended to disinfect dentures. Color constancy of the denture materials is an important parameter to be given due consideration while selecting a disinfectant. This would be a pivotal characteristic for practitioners to consider as they choose disinfectant solutions to disinfect dentures.

Local Defogging Algorithm for Improving Visual Impact in Image Based on Multiobjective Optimization

The preprocessing of images is required for many applications based on industry, social, and academic requirements. Researchers have developed a number of techniques to improve the visual effect of images and appropriately interpret visual effects. The accuracy of visuals is important in cyber security, military organization, police organizations, and forensics to detect the true story from the pictures. They search for evidence by digging deep into the network in search of evidence. If visuals are not clear, preprocessing of images is not done correctly, then it may lead to wrong interpretations. This paper proposes an image local defogging technique based on multiobjective optimization to improve the visual effect of the image as well as the information entropy. The multiobjective function is selected to establish the image reconstruction model based on multiple objectives. The model is utilized to reconstruct a single image to moderate the impact of noise and other interference factors in the original image. The color constancy model and effective detail intensity model are also devised for image enhancement to get the visual details. The atmospheric light value and transmittance are evaluated using a physical model of atmospheric scattering, and the guided filter is used to maximize the transmittance of a single image and improve the efficiency of image defogging. The dark channel priority method is used to realize the local defogging of a single image and to design the local defogging algorithm. Experiments verify the optimization effect of the proposed algorithm in terms of information entropy and container network interface (CNI) value. The tone restoration degree is good, and it improves the overall image quality. The image defogging effect of the proposed algorithm is verified with respect to subjective and objective levels to check the efficacy of the proposed multiobjective model.

DermoCC-GAN: A new approach for standardizing dermatological images using generative adversarial networks

Background and objectiveDermatological images are typically diagnosed based on visual analysis of the skin lesion acquired using a dermoscope. However, the final quality of the acquired image is highly dependent on the illumination conditions during the acquisition phase. This variability in the light source can affect the dermatologist's diagnosis and decrease the accuracy of computer-aided diagnosis systems. Color constancy algorithms have proven to be a powerful tool to address this issue by allowing the standardization of the image illumination source, but the most commonly used algorithms still present some inherent limitations due to assumptions made on the original image. In this work, we propose a novel Dermatological Color Constancy Generative Adversarial Network (DermoCC-GAN) algorithm to overcome the current limitations by formulating the color constancy task as an image-to-image translation problem. MethodsA generative adversarial network was trained with a custom heuristic algorithm that performs well on the training set. The model hence learns the domain transfer task (from original to color standardized image) and is then able to accurately apply the color constancy on test images characterized by different illumination conditions. ResultsThe proposed algorithm outperforms state-of-the-art color constancy algorithms for dermatological images in terms of normalized median intensity and when using the color-normalized images in a deep learning framework for lesion classification (accuracy of the seven-class classifier: 79.2%) and segmentation (dice score: 90.9%). In addition, we validated the proposed approach on two different external datasets with highly satisfactory results. ConclusionsThe novel strategy presented here shows how it is possible to generalize a heuristic method for color constancy for dermatological image analysis by training a GAN. The overall approach presented here can be easily extended to numerous other applications.

Illuminant segmentation for multi-illuminant scenes using latent illumination encoding

Color constancy is an important part of the Human Visual System that allows us to recognize colors of object invariant to the light that is illuminating them. Computational color constancy is the process of estimating the illumination of a scene using some computational method. However, this problem is inherently ill-posed. The RGB value of each pixel is dependent on spectral reflectance of the object and the spectral power distribution of the illumination. Hence, methods that try to solve the computational color constancy problem have to introduce assumptions about the illumination. One common assumption is that there is only one global source of illumination, i.e., that the illumination is constant across the whole scene. Under this assumption, modern color constancy methods achieve excellent results, usually predicting the illumination color with accuracy better than the human eye. This assumption is broken in many real-world multi-illuminant scenes, e.g., outdoor images where parts of the scene are illuminated by either sunlight or skylight. This leads to significant drop in accuracy of single-illuminant estimation methods. Therefore, in this work, we propose a novel method for segmenting images based on illumination in multi-illuminant scenes. This method detects regions where there is only one illuminant, thus detecting areas where the single-illuminant assumption holds. We show that our method produces excellent results and outperforms all baseline models by a large margin.

Transfer Learning for Color Constancy via Statistic Perspective

Color Constancy aims to correct image color casts caused by scene illumination. Recently, although the deep learning approaches have remarkably improved on single-camera data, these models still suffer from the seriously insufficient data problem, resulting in shallow model capacity and degradation in multi-camera settings. In this paper, to alleviate this problem, we present a Transfer Learning Color Constancy (TLCC) method that leverages cross-camera RAW data and massive unlabeled sRGB data to support training. Specifically, TLCC consists of the Statistic Estimation Scheme (SE-Scheme) and Color-Guided Adaption Branch (CGA-Branch). SE-Scheme builds a statistic perspective to map the camera-related illumination labels into camera-agnostic form and produce pseudo labels for sRGB data, which greatly expands data for joint training. Then, CGA-Branch further promotes efficient transfer learning from sRGB to RAW data by extracting color information to regularize the backbone's features adaptively. Experimental results show the TLCC has overcome the data limitation and model degradation, outperforming the state-of-the-art performance on popular benchmarks. Moreover, the experiments also prove the TLCC is capable of learning new scenes information from sRGB data to improve accuracy on the RAW images with similar scenes.