Color Prediction Research Articles

Simulation Calculation of Selective Reflective Films based on Metamaterials and Prediction of Color in Light Filter with Machine Learning

Simulation Calculation of Selective Reflective Films based on Metamaterials and Prediction of Color in Light Filter with Machine Learning

Principal component analysis of refractive index spaces: From glass properties to residual colour prediction

Traditionally glass dispersion is described by Abbe numbers and partial dispersions, and colour correction is explained and visualized e.g. with Pg,F- and Herzberger diagrams. We have recently developed an alternative approach to colour analysis and colour correction based on principal component analysis of normalized refractive index data. The resulting diagrams can be used not only for glass selection, but also for quantitative prediction of partial refractive powers and residual colour aberrations. The method is intrinsically model-free, robust to the choice of dataset and wavelengths, and transfers easily to any spectral range.

NeRC: Rendering Planar Caustics by Learning Implicit Neural Representations.

Caustics are challenging light transport effects for photo-realistic rendering. Photon mapping techniques play a fundamental role in rendering caustics. However, photon mapping methods render single caustics under the stationary light source in a fixed scene view. They require significant storage and computing resources to produce high-quality results. In this paper, we propose efficiently rendering more diverse caustics of a scene with the camera and the light source moving. We present a novel learning-based volume rendering approach with implicit representations for our proposed task. Considering the variety of materials and textures of planar caustic receivers, we decompose the output appearance into two components: the diffuse and specular parts with a probabilistic module. Unlike NeRF, we construct weights for rendering each component from the implicit signed distance function (SDF). Moreover, we introduce the centering calibration and the sine activation function to improve the performance of the color prediction network. Extensive experiments on the synthetic and real-world datasets illustrate that our method achieves much better performance than baselines in the quantitative and qualitative comparison, for rendering caustics in novel views with the dynamic light source. Especially, our method outperforms the baseline on the temporal consistency across frames.

Using Allele-specific PCR to Genotype Single Nucleotide Polymorphisms Associated with Eye Color Prediction

Using Allele-specific PCR to Genotype Single Nucleotide Polymorphisms Associated with Eye Color Prediction

Colour prediction as a weaving design selection help tool in Jacquard CAD

The final colour prediction of a weave design made of dyed yarns is a difficult problem. This study shows how a geometric model can be developed to obtain the final colour prediction objectively. For this purpose, a woven material was divided into weft, warp and pores. Then, all parameters needed for the calculation of each colour contribution were identified. A geometrical model based on construction parameters was developed to predict the surface colour contribution of each coloured yarn in a weave surface. To validate the predicted colorimetric data, a visual assessment experiment was conducted. Then, the difference between the predicted and actual colour appearance of the weave pattern was evaluated and analysed in function of weaving structures, and weft yarns colours. For this purpose, simple woven structures (plain, twill 1/3, basket 2/2 and satin Turc) with four coloured weft yarns were used. Results show that the proposed model could correctly predict the final colour of weave designs. Therefore, the model has the potential to eliminate subjective evaluations and reduce prototype sample production by automating the process of weave/colour simulation, thereby reducing the cost and time for product development. The methods of utilization of colour in woven textiles depend upon the composition of the weave design to be woven and the structure parameters of the cloth.

A Method of Enhancing Silk Digital Printing Color Prediction through Pix2Pix GAN-Based Approaches

Color prediction and color management for digital printed fabrics remain a challenging task. Accurate prediction of color appearances of digital printed fabrics would enable designers and manufacturers to better fulfill their design requirements and creative visions. We propose a color prediction method for silk digital printing utilizing a Pix2Pix Generative Adversarial Network (GAN) framework. This method aims to generate predicted images that possess the same stylistic and color characteristics as their actual fabrics after production. To develop and validate the method, color data and images are collected and processed from 5252 sets of paired original Pantone TPX color card and actual print sample fabrics. The results of this study demonstrate that the method can predict the colors of silk digital print samples while effectively reproducing the effects of inkjet printing in silk fabrics including silk crepe satin and silk twill. The method exhibits high prediction accuracy to an average CIEDE2000 value of 2.372 for silk crepe satin and 1.846 for silk twill. The findings of this research not only enhance the efficiency and accuracy of color management in fabric digital printing technology but also contribute to the exploration and development of high-fidelity color prediction techniques within the textile industry.

Connection between SDSS galaxies and ELUCID subhaloes in the eye of machine learning

ABSTRACT We explore the feasibility of learning the connection between Sloan Digital Sky Survey (SDSS) galaxies and ELUCID subhaloes with random forest (RF). ELUCID is a constrained N-body simulation constructed using the matter density field of SDSS. Based on a SDSS-ELUCID matched catalogue, we build RF models that predict Mr magnitude, colour, stellar mass M*, and specific star formation rate (sSFR) with several subhalo properties. While the RF can predict Mr and M* with reasonable accuracy, the prediction accuracy of colour and sSFR is low, which could be due to the mismatch between galaxies and subhaloes. To test this, we shuffle the galaxies in subhaloes of narrow mass bins in the local neighbourhood using galaxies of a semi-analytic model (SAM) and the TNG hydrodynamic simulation. We find that the shuffling only slightly reduces the colour prediction accuracy in SAM and TNG, which is still considerably higher than that of the SDSS. This suggests that the true connection between SDSS colour and subhalo properties could be weaker than that in the SAM and TNG without the mismatch effect. We also measure the Pearson correlation coefficient between the galaxy and subhalo properties in SDSS, SAM, and TNG. Similar to the RF results, we find that the colour–subhalo correlation in SDSS is lower than in both SAM and TNG. We also show that the galaxy–subhalo correlations depend on subhalo mass in the galaxy formation models. Advanced surveys with fainter galaxies will provide new insights into the galaxy–subhalo relation in the real Universe.

Prediction of the Structural Color of Liquid Crystals via Machine Learning

Materials that generate structural color may be promising alternatives to dyes and pigments due to their relative long-term stability and environmentally benign properties. Liquid crystal (LC) mixtures of cholesteryl esters demonstrate structural color due to light reflected from the helical structure of the self-assembled molecules. The apparent color depends on the pitch length of the liquid crystal. While a wide range of colors have been achieved with such LC formulations, the nature of the pitch–concentration relationship has been difficult to define. In this work, various machine learning approaches to predict the reflected wavelength, i.e., the position of the selective reflection band, based on LC composition are compared to a Scheffe cubic model. The neural network regression model had a higher root mean squared error (RMSE) than the Scheffe cubic model with improved predictions for formulations not included in the dataset. Decision tree regression provided the best overall performance with the lowest RMSE and predicted position of the selective reflection band within 0.8% of the measured values for LC formulations not included in the dataset. The predicted values using the decision tree were over two-fold more accurate than the Scheffe cubic model. These results demonstrate the utility of machine learning models for predicting physical properties of LC formulations.

Dual-color space network with global priors for photo retouching

There have been growing trends using deep learning-based approaches for photo retouching which aims to enhance unattractive images and make them visually appealing. However, the existing methods only considered the RGB color space, which limited the available color information for editing. To address this issue, we propose a dual-color space network that extracts color representations from multiple color spaces to provide more robust color information. Our approach is based on the observation that converting an image to a different color space generates a new image that can be further processed by a neural network. Hence, we utilize two separate networks: a transitional network and a base network, each operating in a different color space. Specifically, the input RGB image is converted to another color space (e.g., YCbCr) using color space converter (CSC). The resulting image is then passed through the transitional network to extract color representations from the corresponding color space using color prediction module (CPM). The output of the transitional network is converted back to the RGB space and fed into the base network, which operates in RGB space. By utilizing global priors from each representation in different color spaces, we guide the retouching process to produce natural and realistic results. Experimental results demonstrate that our proposed method outperforms state-of-the-art methods on the MIT-Adobe FiveK dataset, and an in-depth analysis and ablation study highlight the advantages of our approach.

Eye and hair color prediction of human DNA recovered from Lucilia sericata larvae.

Forensic entomological evidence is employed to estimate minimum postmortem interval (PMImin), location, and identification of fly samples or human remains. Traditional forensic DNA analysis (i.e., STR, mitochondrial DNA) has been used for human identification from the larval gut contents. Forensic DNA phenotyping (FDP), predicting human appearance from DNA-based crime scene evidence, has become an established approach in forensic genetics in the past years. In this study, we aimed to recover human DNA from Lucilia sericata (Meigen 1826) (Diptera: Calliphoridae) gut contents and predict the eye and hair color of individuals using the HIrisPlex system. Lucilia sericata larvae and reference blood samples were collected from 30 human volunteers who were under maggot debridement therapy. The human DNA was extracted from the crop contents and quantified. HIrisPlex multiplex analysis was performed using the SNaPshot minisequencing procedure. The HIrisPlex online tool was used to assess the prediction of the eye and hair color of the larval and reference samples. We successfully genotyped 25 out of 30 larval samples, and the most SNP genotypes (87.13%) matched those of reference samples, though some alleles were dropped out, producing partial profiles. The prediction of the eye colors was accurate in 17 out of 25 larval samples, and only one sample was misclassified. Fourteen out of 25 larval samples were correctly predicted for hair color, and eight were misclassified. This study shows that SNP analysis of L. sericata gut contents can be used to predict eye and hair color of a corpse.

Research on a colour solid built by gridded colour mixing of nine primary‐coloured fibres and its neural network colour prediction approach

AbstractAccording to the demand for colour prediction for coloured yarn, two adjacent colours chosen from red (R), yellow (Y), green (G), cyan (C), blue (B) and magenta (M) fibres were combined with fibres of dark grey (O1), medium grey (O2) and light grey (O3), respectively, and then ternary coupling‐superposition mixing was performed to acquire a colour solid consisting of three lightnesses, 18 colour mixing units and 18 × (m + 1) × n grid points. An integrated colour mixing with 20% hue gradient and 33.33% saturation gradient was performed to achieve a colour solid containing 360 grid points, then using it as the sample space for the colour prediction model. A total of 360 typical samples were established by the grid points, 213 yarns and fabrics were prepared by the typical sample parameters, and the corresponding reflectance was accessed by a spectrophotometer. Neural network models for predicting reflectance by mixing ratios as well as forecasting mixing ratios by reflectance, were established. The 12 non‐grid point parameters were chosen to prepare corresponding yarns and fabrics, and the corresponding reflectance was measured. The predicted and measured values of the neural network model were compared to verify its predictive ability and generalisability. The results showed that when predicting the colour by the mixing ratios, the colour difference between the predicted and measured samples ranged from 1.5 to 3.4, with an average of 2.4; and when forecasting the mixing ratios by the colour, the colour difference ranged from 0.8 to 5.6, with an average of 2.4.

A Stearns–Noechel colour prediction model reconstructed from gridded colour solid of nine primary colours and its application

AbstractA full gamut colour solid model consisting of three lightness planes, 18 colour mixing units and 360 grid points is constructed from nine primary coloured fibres: red (R), yellow (Y), green (G), cyan (C), blue (B), magenta (M), dark grey (O1), medium grey (O2) and light grey (O3). Subsequently, the 213 coloured yarns and fabrics containing different lightness, hue and saturation were prepared according to the mixing ratio parameters in the colour solid. The Stearns–Noechel colour prediction algorithm, which predicts reflectance using coloured fibre mixing ratios, was improved and applied according to the requirements of colour prediction; and the Stearns–Noechel proportion prediction algorithm, which predicts coloured fibre mixing ratios by reflectance, was refined and employed in accordance with the demands of proportion prediction. Then, the 12 additional coloured fabrics were fabricated and their corresponding measurement data were used on the algorithm for validating its forecasting capabilities. The final experimental results reveal that the maximum colour difference for colour prediction is 5.5, the minimum is 1.7, and the average is 3.7; the maximum colour difference for proportion prediction is 3.3, the minimum is 0.3, and the average is 1.6. Therefore, this approach is promising to improve the colour reproduction issues encountered in the processing of three‐channel computer numerical control (CNC) spinning.

Text2Color Networks: Deep Learning Models for Color Generation from Compositional Color Descriptions

Color serves as an important cue in graphics, arts and in many computer vision applications. We reliably and effortlessly name colors and communicate them with their names. However, many applications such as graphics, color design, palette generation and color selection tools demand numerical values of colors. Predicting and communicating colors by their numerical values is less intuitive and difficult task as it is a mapping of millions of colors for a specific color space. To bridge the gap between linguistic color names and numerical color values, in this paper, we present neural network architectures that predict a point in color space for a given color name. Proposed models provide user interface between colors and names by learning the language semantics and mimic human level comprehension of color descriptions to predict colors and modify them with respect to linguistic adjectives of color names. We consider color prediction as a regression problem and solve it as a language modeling task. Color descriptions are taken as text sequences and each sentence is represented with word-level tokenization. Each token is transformed into a word vector in the latent space using CBOW word embeddings model. Word vectors representing color names are fed as input to neural networks and trained with normalized R, G, B values as supervision information. Trained models are capable of predicting color for a given color name and modify colors for different nouns and adjectives associated with color names. We also built color generation models based on pre-trained word embeddings to overcome the limited availability of large linguistic color name datasets. These pre-trained models perform well with datasets containing few thousand color names. We then present two recommendation engines that suggest similar color palette to user given color name. These recommendation engines enhance the color vocabulary and assist users in the color selection process.

Forensic DNA Phenotyping: Genes and Genetic Variants for Eye Color Prediction.

In recent decades, the use of genetic polymorphisms related to specific phenotypes, such as eye color, has greatly contributed to the development of the research field called forensic DNA phenotyping (FDP), enabling the investigators of crime cases to reduce the number of suspects, making their work faster and more precise. Eye color is a polygenic phenotype, and many genetic variants have been highlighted, with the major contributor being the HERC2-OCA2 locus, where many single nucleotide variations (SNPs) were identified. Interestingly, the HERC2-OCA2 locus, containing the intronic SNP rs12913832, the major eye color determinant, shows a high level of evolutionary conservation across many species of vertebrates. Currently, there are some genetic panels to predict eye color by genomic DNA analysis, even if the exact role of the SNP variants in the formation of eye color is still poorly understood, with a low level of predictivity in the so-called intermediate eye color. Many variants in OCA2, HERC2, and other genes lie in introns or correspond to synonymous variants, highlighting greater complexity in the mechanism of action of such genes than a simple missense variation. Here, we show the main genes involved in oculocutaneous pigmentation and their structural and functional features, as well as which genetic variants show the highest level of eye color predictivity in currently used FDP assays. Despite the great recent advances and impact of FDP in criminal cases, it is necessary to enhance scientific research to better understand the mechanism of action behind each genetic variant involved in eye color, with the goal of obtaining higher levels of prediction.

Language-based Photo Color Adjustment for Graphic Designs

Adjusting the photo color to associate with some design elements is an essential way for a graphic design to effectively deliver its message and make it aesthetically pleasing. However, existing tools and previous works face a dilemma between the ease of use and level of expressiveness. To this end, we introduce an interactive language-based approach for photo recoloring, which provides an intuitive system that can assist both experts and novices on graphic design. Given a graphic design containing a photo that needs to be recolored, our model can predict the source colors and the target regions, and then recolor the target regions with the source colors based on the given language-based instruction. The multi-granularity of the instruction allows diverse user intentions. The proposed novel task faces several unique challenges, including: 1) color accuracy for recoloring with exactly the same color from the target design element as specified by the user; 2) multi-granularity instructions for parsing instructions correctly to generate a specific result or multiple plausible ones; and 3) locality for recoloring in semantically meaningful local regions to preserve original image semantics. To address these challenges, we propose a model called LangRecol with two main components: the language-based source color prediction module and the semantic-palette-based photo recoloring module. We also introduce an approach for generating a synthetic graphic design dataset with instructions to enable model training. We evaluate our model via extensive experiments and user studies. We also discuss several practical applications, showing the effectiveness and practicality of our approach. Please find the code and data at https://zhenwwang.github.io/langrecol.

A Multisample Approach in Forensic Phenotyping of Chronological Old Skeletal Remains Using Massive Parallel Sequencing (MPS) Technology.

It is very important to generate phenotypic results that are reliable when processing chronological old skeletal remains for cases involving the identification of missing persons. To improve the success of pigmentation prediction in Second World War victims, three bones from each of the eight skeletons analyzed were included in the study, which makes it possible to generate a consensus profile. The PowerQuant System was used for quantification, the ESI 17 Fast System was used for STR typing, and a customized version of the HIrisPlex panel was used for PCR-MPS. The HID Ion Chef Instrument was used for library preparation and templating. Sequencing was performed with the Ion GeneStudio S5 System. Identical full profiles and identical hair and eye color predictions were achieved from three bones analyzed per skeleton. Blue eye color was predicted in five skeletons and brown in three skeletons. Blond hair color was predicted in one skeleton, blond to dark blond in three skeletons, brown to dark brown in two skeletons, and dark brown to black in two skeletons. The reproducibility and reliability of the results proved the multisample analysis method to be beneficial for phenotyping chronological old skeletons because differences in DNA yields in different bone types provide a greater possibility of obtaining a better-quality consensus profile.

DDGAN: Dense Residual Module and Dual-stream Attention-Guided Generative Adversarial Network for colorizing near-infrared images

Transforming near-infrared(NIR) images into realistic RGB images is a challenging task. Recently, with the development of deep learning, the colorization of NIR images has been significantly improved. However, there are still problems with distorted textures and blurred details. The main reason is that NIR images require color prediction and gray-scale detail recovery to be colorized. Moreover, NIR images contain limited detail and lack color information, which poses a serious challenge for feature extraction. To address the problems, we propose the Dense Residual Module and Dual-stream Attention-Guided Generative Adversarial Network (DDGAN) in this paper. The Dense Residual Module (DRM) improves the network feature extraction capability in the form of dense residuals and increases the network depth to improve the adaptability of the network. The Dual-stream Attention Module (DAM) further improves the quality of colorized images by enhancing important features and suppressing unnecessary features to focus on essential visual features. We propose a composite loss function consisting of content loss, adversarial loss, perceptual loss, synthesized loss, and total variation loss, which improves the quality of colorful images in terms of both edge structure and visual perception. We consider the NIR images for the visible image conversion task to evaluate the efficiency and performance of the proposed model. The proposed DDGAN outperforms most existing methods in terms of efficiency and quality of the generated images on the RGB–NIR dataset and the OMSIV dataset. Compared to state-of-the-art methods, the proposed DDGAN shows promising results with significant improvements in PSNR, SSIM, MSE, and NRMSE. Extensive experimental data show that the proposed DDGAN can produce state-of-the-art colorized NIR images in objective metrics and subjective quality.

Deep Learning-Based Machine Color Emotion Generation

This paper investigates generating machine color emotion through deep learning. The grayscale image colorization model's training process resembles human memory color. Sixty images were recolored and quality evaluated to explore machine generated color impressions. Six experimental samples were recolored under D65, A, CWF, and TL84 light sources. Changes in lightness, chroma, and hue angle compared the original and colorized images, exploring light source effects on machine color perception. Analyzing differences in coloring results within the CIEL* a* b* color space for pixels with equal grayscale verified machine color emotion generation. Results show the machine learns to form color impressions from samples. Different light source color temperatures impact color prediction accuracy. The machine accurately colors images based on semantic context, demonstrating spontaneous color emotion generation through deep learning. This research positively contributes to the development of intelligent devices with color emotion.

Construction of Color Prediction Model for Damaged Korla Pears during Storage Period

Construction of Color Prediction Model for Damaged Korla Pears during Storage Period

Analysis of beef quality according to color changes using computer vision and white-box machine learning techniques

The quality of beef products relies on the presence of a cherry red color, as any deviation toward brownish tones indicates a loss in quality. Existing studies typically analyze individual color channels separately, establishing acceptable ranges. In contrast, our proposed approach involves conducting a multivariate analysis of beef color changes using white-box machine learning techniques. Our proposal encompasses three phases. (1) We employed a Computer Vision System (CVS) to capture the color of beef pieces, implementing a color correction pre-processing step within a specially designed cabin. (2) We examined the differences among three color spaces (RGB, HSV, and CIELab*) (3) We evaluated the performance of three white-box classifiers (decision tree, logistic regression, and multivariate normal distributions) for predicting color in both fresh and non-fresh beef. These models demonstrated high accuracy and enabled a comprehensive understanding of the prediction process. Our results affirm that conducting a multivariate analysis yields superior beef color prediction outcomes compared to the conventional practice of analyzing each channel independently.