Search

Introduction: Face Detection is used in many different steams like video conferencing, human-computer interface, face detection, and in the database management of image. Therefore, the aim of our paper is to apply Red Green Blue (RGB) and Hue Saturation Value (HSV) color models in detecting the single, including multiple images of any face. Each color model HSV, Ycbcr and TSL are individually performed with the RGB color model to detect from the single and multiple images, the region of the face. Methods: The morphological operations are performed in the face region to a number of pixels as the proposed parameter to check either an input image contains face region or not. Canny edge detection is also used to show the boundaries of a candidate face region, in the end, the face can be shown detected by using bounding box around the face. Results: The reliability model has also been proposed for detecting the faces in single and multiple images. The results of the experiments reflect that the algorithm been proposed performs very well in each model for detecting the faces in single and multiple images and the reliability model provides the best fit by analyzing the precision and accuracy. Moreover Discussion: The calculated results show that HSV model works best for single faced images whereas YCbCr and TSL models work best for multiple faced images. Also, the evaluated results by this paper provides the better testing strategies that helps to develop new techniques which leads to an increase in research effectiveness. Conclusion: The calculated value of all parameters is helpful for proving that the proposed algorithm has been performed very well in each model for detecting the face by using a bounding box around the face in single as well as multiple images. The precision and accuracy of all three models are analyzed through the reliability model. The comparison calculated in this paper reflects that HSV model works best for single faced images whereas YCbCr and TSL models work best for multiple faced images.

The paper deals with the issues of detection and modeling of human faces and objects on a face taken from the images. The model, algorithm and program are developed for the detection of human facial contours and main elements. The preliminary image processing involves the methods of color modeling and color measurements. Well-known methods, including hidden Markov models, are used for image recognition and processing. The training of the developed model was carried out with neural network methods of machine learning based on a specially created sample, as well as using color segmentation methods. A factor model of a human face is created, which makes it possible to select and recognize efficiently a face and its objects in the image at high speed and with a given accuracy. The experiments have shown that the accuracy of the correct selection of boundaries was about 95–96 % after training. The developed model can be used in security assurance tasks, namely to search and identify criminals, to strengthen law and order, to control access to critical infrastructure facilities, etc.

Given a pair of images---target person and garment on another person---we automatically generate the target person in the given garment. Previous methods mostly focused on texture transfer via paired data training, while overlooking body shape deformations, skin color, and seamless blending of garment with the person. This work focuses on those three components, while also not requiring paired data training. We designed a pose conditioned StyleGAN2 architecture with a clothing segmentation branch that is trained on images of people wearing garments. Once trained, we propose a new layered latent space interpolation method that allows us to preserve and synthesize skin color and target body shape while transferring the garment from a different person. We demonstrate results on high resolution 512 × 512 images, and extensively compare to state of the art in try-on on both latent space generated and real images.

We propose a layout estimation method for multi-layered ink using a measurement of the line spread function (LSF) and machine learning. The three-dimensional printing market for general consumers focuses on the reproduction of realistic appearance. In particular, for the reproduction of human skin, it is important to control translucency by adopting a multilayer structure. Traditionally, layer design has depended on the experience of designers. We, therefore, developed an efficient layout estimation to provide arbitrary skin color and translucency. In our method, we create multi-layered color patches of human skin and measure the LSF as a metric of translucency, and we employ a neural network trained with the data to estimate the layout. As an evaluation, we measured the LSF from the computer-graphics-created skin and fabricate skin using the estimated layout; evaluation with root-mean-square error showed that we can obtain color and translucency that are close to the target.

We propose a non-contact heart rate (HR) estimation method that is robust to various situations, such as bright, low-light, and varying illumination scenes. We utilize a camera that records red, green, and blue (RGB) and near-infrared (NIR) information to capture the subtle skin color changes induced by the cardiac pulse of a person. The key novelty of our method is the adaptive fusion of RGB and NIR signals for HR estimation based on the analysis of background illumination variations. RGB signals are suitable indicators for HR estimation in bright scenes. Conversely, NIR signals are more reliable than RGB signals in scenes with more complex illumination, as they can be captured independently of the changes in background illumination. By measuring the correlations between the lights reflected from the background and facial regions, we adaptively utilize RGB and NIR observations for HR estimation. The experiments demonstrate the effectiveness of the proposed method.

This paper integrates skin color model and improved AdaBoost into a face detection method for high-resolution images with complex backgrounds. Firstly, the skin color areas were detected in a multi-color space. Each image was subject to adaptive brightness compensation, and converted into the YCbCr space, and a skin color model was established to solve face similarity. After eliminating the background interference by morphological method, the skin color areas were segmented to obtain the candidate face areas. Next, the inertia weight control factors and random search factor were introduced to optimize the global search ability of particle swarm optimization (PSO). The improved PSO was adopted to optimize the initial connection weights and output thresholds of the neural network. After that, a strong AdaBoost classifier was designed based on optimized weak BPNN classifiers, and the weight distribution strategy of AdaBoost was further improved. Finally, the improved AdaBoost was employed to detect the final face areas among the candidate areas. Simulation results show that our face detection method achieved high detection rate at a fast speed, and lowered false detection rate and missed detection rate.

Tujuan dari deteksi objek manusia adalah untuk mengidentifikasi objek manusia dalam image. Adanya perubahan objek manusia yang dipengaruhi oleh faktor-faktor internal seperti ekspresi wajah, bentuk tubuh, warna kulit, dan pergerakan badan serta dipengaruhi oleh faktor eksternal seperti kondisi pencahayaan, latar belakang yang bervariasi dan sudut pengambilan image. Mendeteksi objek manusia terhadap beberapa kondisi seperti didapatkan pada image yang berisi lebih dari satu objek, dan ukuran objek yang bervariasi didalam image. Penelitian ini bertujuan untuk menganalisis dan merancang aplikaasi untuk mendeteksi objek manusia yang ada dalam image. Algoritma thinning digunakan untuk pengenalan pola objek manusia dari satu image untuk gambar yang lain agar menyamakan objek piksel untuk pendeteksian yang maksimal. Local maxima digunakan sebagai fungsi untuk memaksimumkan pencarian piksel pada objek dalam image untuk mempermudah dan mempercepat deteksi image.

Purpose Maxillofacial prosthetic rehabilitation replaces missing structures to recover the function and aesthetics relating to facial defects or injuries. Deep learning is rapidly expanding with respect to applications in medical fields. In this study, we apply the artificial neural network (ANN)-based deep learning approach to coloration support for fabricating maxillofacial prostheses. Methods We compared two machine learning algorithms, ANN-based deep learning and the random forest algorithm, to determine the compounding amount of pigment. We prepared 52 silicone elastomer specimens of varying colors and measured the CIE 1976 L* a* b* color space information using a spectrophotometer on the input dataset. The output of these algorithms indicated the compounding amount of four pigments. According to the algorithms’ pigment compounding predictions, we prepared the specimens for validation analysis and measured the CIE 1976 L* a* b* values. We determined the color differences between the real skin color of five research participants (22.3 ± 1.7 years) and that of the silicone elastomer specimens fabricated based on the algorithm predictions using the CIEDE00 ΔE00 color system. Results The color differences (ΔE00 value) between the real skin color and silicone elastomer validation specimens were 3.45 ± 0.87 (ANN) and 5.54 ± 1.41 (random forest), which indicates that the deep ANN approach produced superior results with respect to the ΔE00 value compared with the random forest algorithm. Conclusions These results suggest that applying deep ANN is a promising technique for the coloration of maxillofacial prostheses.

Proses pematangan tandan buah segar (TBS) kelapa sawit dapat dilihat dari perubahan warna kulit buahnya. Hanya TBS yang layak untuk diolah yang akan diangkut truk untuk dibawa ke pabrik. Akan tetapi, untuk saat ini TBS yang tidak layak angkut tetap terangkut ke pabrik. Penelitian ini menghasilkan aplikasi berbasis web yang dikelola admin untuk mengelola data training dan aplikasi berbasis android yang dapat memprediksi TBS yang layak dan tidak layak angkut ke pabrik. Petani cukup memfoto TBS, kemudian aplikasi akan memprediksi label. Hasil prediksi yang didapatkan berupa TBS kelapa sawit layak angkut dan tidak layak angkut. Ekstraksi ciri yang digunakan adalah warna RGB, nilai numerik yang didapat dari ekstraksi ciri kemudian diolah dengan algoritma. Algoritma yang digunakan pada aplikasi ini adalah k-nearest neighbor. Berdasarkan pengujian yang dilakukan, aplikasi memiliki tingkat akurasi mencapai 92% dengan nilai K=7.

An animal’s pelage, feather, or skin color can serve a variety of functions, so it is important to have multiple standardized methods for measuring color. One of the most common and reliable methods for measuring animal coloration is the use of standardized digital photographs of animals. New technology in the form of a commercially available handheld digital color sensor could provide an alternative to photography-based animal color measurements. To determine whether a digital color sensor could be used to measure animal coloration, we tested the ability of a digital color sensor to measure coloration of mammalian, avian, and lepidopteran museums specimens. We compared results from the sensor to measurements taken using traditional photography methods. Our study yielded significant differences between photography-based and digital color sensor measurements of brightness (light to dark) and colors along the green to red spectrum. There was no difference between photographs and the digital color sensor measurements for colors along the blue to yellow spectrum. The average difference in recorded color (ΔE) by the 2 methods was above the threshold at which humans can perceive a difference. There were significant correlations between the sensor and photographs for all measurements indicating that the sensor is an effective animal coloration measuring tool. However, the sensor’s small aperture and narrow light spectrum range designed for human-vision limit its value for ecological research. We discuss the conditions in which a digital color sensor can be an effective tool for measuring animal coloration in both laboratory settings and in the field.

Apple is one of the popular fruits for public consumption. People can distinguish many apples based on their colors and shapes, such as the Braeburn Apple with skin color varies from orange to red, the Pink Lady Apple that is red with pseudo pink, the Crismon Snow Apple that has dark red skin. Recently, computers can automatically recognize them using digital image processing techniques such as Convolutional Neural Networks (CNN). In this paper, a CNN-based classification model of apple types is developed using 1856 apple images from three classes derived from the fruit-360 dataset on the Kaggle website, and its robustness is then examined. Two types of testing have been carried out in this study: testing five scenarios for sharing training data and testing five scenarios for robustness to noise. An examination based on 5-fold cross-validation shows that CNN is robust to decreasing the portion of training set size up to 50% to get high accuracy of 99.97% in classifying 50% testing set, which is better than previous models that use VGG16, faster R-CNN, and Tanh. Decreasing the portion training set to 40% and 30% reduces the accuracy to 95.97% and 95.29%, respectively. Adding low-level noises of 10% into the testing images decreases the accuracy slightly to 99.17%. However, high-level noises of 50% drastically make the accuracy drastically drops to 63.93%.

Real-Time Vein Imaging is a device with which we aim to locate the veins in our arms easily for any medical purpose for intravenous (IV) access. Most electronic vein viewers are very expensive or complex in design. Our work presents an affordable, simple designed, easy to use portable device which can overcome the difficulties faced by phlebotomists or doctors during vein puncture. This device can be used irrespective of age, skin color, thickness or collapsed veins. The design of our prototype has been developed to capture the vein images using multispectral near infrared (NIR) source. The veins can be seen and traced on the skin on exposure of NIR making it easy to view and select the right vein and not cause patients unnecessary pain due to multiple punctures. Utilization of this device could help the jobs of the doctors and phlebotomists of developing countries such as Bangladesh.

Background Artificial intelligence (AI) image recognition models have been relatively successful in diagnosing cutaneous manifestations in individuals with light skin tone. However, when these models are tested on the same cutaneous manifestations in individuals with darker or brown skin tone, the performance of the model drops due to a paucity of such images available for model training. Objective The objective of this study was to improve the performance of AI models in recognizing cutaneous diseases in individuals with darker skin tone. Methods Unsupervised computer darkening of skin color with preservation of the dermatological disease/lesion characteristics in images of light skinned individuals with basal cell carcinoma (BCC) and melanoma was performed. Results Training an AI model on these artificially "darkened" images as compared to training on the original light skinned images resulted in a higher sensitivity, specificity, positive predictive value, negative predictive value, F1 score and area under the receiver operating characteristic curve of the AI model in differentiating between BCC and melanoma in individuals with brown skin tone. Conclusion Use of unsupervised image to image translation in medical AI image recognition models has the potential to significantly improve their accuracy in diagnosing diseases in individuals with racially diverse skin tone.

The conventional skin color separation method suffers from the severe limitation that the input skin image must have no shading component and must have substantial variation in the pigment density distribution, simultaneously. We therefore propose an improved method, which uses two input skin images instead of one. In this improved method, one input image has no shading component and the other has substantial variation in pigment density distribution. To verify the effect of our method, we took images of four subjects and separated them into pigment components. We then compared the independence evaluation value of the separated signals extracted using our method and that of the conventional method. The results show that we obtain separated signals that have a better independence evaluation value by our method than by the conventional method.

Summary In response to the changing environment, chameleons can adjust their skin colors to communicate and create disguise during locomotions. By mimicking their behaviors, we present structural-color actuators that can sense the environment and give responses of non-fatigued vivid color alterations and programmable shape transformations. Due to the sensitive vapochromic and robust vapomechanical response properties, these structural-color actuators exhibit fast color changes (less than 1 s) and non-fatigued shape transformations (more than 110 cycles) when exposed to specific vapors. On the basis of these extraordinary properties, we demonstrate the actuation processes of a rotating pinwheel and a closing/blooming flower with brilliant color changes, and a worm-like structural-color walker that can move forward straight and rhythmically with dynamic color changes. With these chameleon-like behaviors, these structural-color actuators could have important implications in sensing, communication, and disguise of soft robotics.

In this study, we analyzed color changed by wetness in various areas on palm to clarify the influence of highlights. To improve the accuracy of emotion estimation, it is necessary to suggest other modalities combined with the conventional modalities. Therefore, we focused on emotional sweating, which is reliable modality of contact methods but not used in non-contact methods. We analyzed the color change in images containing a few highlights to clarify the influence of sharp highlights because the highlights can be noise for the analysis of internal reflection change by getting wet. The sharp highlight means the bright pixels in images of gloss material. We also analyzed areas separated from pixels representing sharp highlights to unveil the influence of the sharp highlights. As a result, we found that sharp highlights have a little influence on the analysis and better detection can be performed on pixels not containing sharp highlight area.

Face detection (face detection) is one of the initial steps that is very important before the face recognition process (face recognition). Face detection is the detection of objects in the form of faces in which there are special features that represent the shape of faces in general. One method of face detection is the Viola Jones method. Viola Jones method is used to detect faces and skin color segmentation, test data processing using Matlab and capture on a Smartphone. The test is carried out at normal light intensity with a predetermined distance and face position. The results of this study indicate the level of accuracy of detection of face image variations in the position of face images facing forward (frontal), sideways left and right 45̊. But it has a weakness of this face detection system that is unable to determine faces in images that have faces that are not upright (tilted) or not frontal (facing sideways) at a 90̊ angle. Face position that is upright / not upright will determine the success of this face detection. The level of identification of the Viola Jones simulation was 100% with 4 images consisting of 3 boys and 1 girl.

A robust skin color detection method using adaptive skin color modeling on YES color space is proposed. The skin colors are classified into three skin regions according to the Y value, and each reg...

How does the accuracy of deep neural network models trained to classify clinical images of skin conditions vary across skin color? While recent studies demonstrate computer vision models can serve as a useful decision support tool in healthcare and provide dermatologist-level classification on a number of specific tasks, darker skin is under-represented in the data. Most publicly available data sets do not include Fitzpatrick skin type labels. We annotate 16,577 clinical images sourced from two dermatology atlases with Fitzpatrick skin type labels and open-source these annotations. Based on these labels, we find that there are significantly more images of light skin types than dark skin types in this dataset. We train a deep neural network model to classify 114 skin conditions and find that the model is most accurate on skin types similar to those it was trained on. In addition, we evaluate how an algorithmic approach to identifying skin tones, individual typology angle, compares with Fitzpatrick skin type labels annotated by a team of human labelers.

Skin color target detection is a typical single-type target detection. However, the existing multi-targets detection networks take a long training time and have complex structure for skin color target detection. The detection speeds of some network can be further improved. This paper designs a simple structure for rapid skin color model detection by combining RPN network and knowledge of skin color ellipse model. According to the pretreatment of VGG-16, the test is executed in Kitchen-l in behavior data set Watch-n-patch and Asian face data set CASIA-FaceV5. The frame rate reaches 10.2 and 11.3, and the average precision (AP) reaches to 77.8% and 91.2%, respectively. Further, the desired results are obtained under the video and camera data.

Designing and developing autonomous vehicles that are capable of moving safely on roads by sensing the environment has motivated researchers to focus on pedestrian detection systems so they can detect people as fast and accurately as possible. However, for pedestrian detection, it is crucial to consider not only the pedestrians themselves but their color as well, because color has the advantage of being invariant to changes in scaling, rotation, and partial occlusion. Therefore, considering skin color detection for implementing pedestrian detection systems is an essential required step in ensuring autonomous vehicles are further incorporated into our society. Detecting human skin has proven to be a challenging problem because skin color can vary dramatically in its appearance due to many factors such as illumination, race, imaging conditions, and others. Recently it has been noted that pedestrian detection systems for autonomous vehicles perform poorly at detecting people with darker skin tones. Such findings indicate that there is a larger problem that is causing these issues: algorithmic bias. Algorithmic bias in pedestrian detection systems could be the leading factor of their poor performance due to the methods implemented and datasets used. Unfortunately, the algorithmic bias in this context has not been considered closely and it seems that many studies do not cover this aspect closely when discussing pedestrian detection systems for autonomous vehicles. To alleviate this, we attempt to explore different techniques that can be used to detect pedestrians while minimizing bias. In our experiment, we use both a YOLO v3 convolutional neural network and K-Means clustering for classifying skin-tones.

Melanoma is one of the most fatal forms of skin cancer and is often very difficult to differentiate from other benign skin lesions. However, if detected at its early stages, it can almost always be cured. Researchers and data scientists have studied this disease in-depth with the help of large datasets containing high-quality dermascopic images, such as those assembled by the International Skin Imaging Collaboration (ISIC). However, these images often lack diversity and over-represent patients with very common skin features such as light skin and having no visible body hair. In this study, we introduce a novel architecture called LatentNet which automatically detects over-represented features and reduces their weights during training. We tested our model on four distinct categories - three skin color levels corresponding to Type I, II, and III on the Fitzpatrick Scale, and images containing visible hair. We then compared the accuracy against the conventional Deep Convolutional Neural Network (DCNN) model trained using the standard approach (i.e. without detecting over-represented features) and containing the same hyper-parameters as the LatentNet. LatentNet showed significant performance improvement over the standard DCNN model with accuracy of 89.52%, 79.05%, 64.31%, and 64.35% compared to the DCNN accuracy of 90.41%, 70.82%, 45.28%, 56.52% in the corresponding categories, respectively. Differences in the average performance between the models were statistically significant (p < 0.05), suggesting that the proposed model successfully reduced bias amongst the tested categories. LatentNet is the first architecture that addresses racial bias (and other sources of bias) in deep-learning based Melanoma diagnosis.

Detecting skin color changes has medical diagnostic value. Spectral optimization of a 16-channel LED for cyanosis observation index, circadian, color, and efficacy metrics underline the benefits of using tunable lighting systems in healthcare facilities.

Face attributes play an important role in face-related applications. However, existing face attributes (such as expression, age, and skin color) are subject to change. The ethnicity attribute is precious due to its invariance over time, but has not been developed well. This is partly because there is no large enough dataset and labeled accurately with ethnicity attribute. This paper proposes a new Face Dataset with Ethnicity Attribute (FDEA), intended for ethnicity recognition benchmark. For this purpose, we first collect an initial face dataset from CelebA and LFWA [10], MORPH [13], UTKFace [20], FairFace [8], and the web. The samples extracted from CelebA are not labeled with ethnicity attribute. To this end, we employ nine annotators to label these samples from CelebA, while cleaning the remaining samples manually. Finally, our FDEA contains 157,801 samples and is divided into three classes: Caucasian (54,438), Asian (61,522), and African (41,841). Moreover, we carry out a benchmark experiment by testing eight mainstream backbones on the proposed FDEA. The baseline results of the three-classification accuracy are all over 0.92. FDEA is publicly available at https://github.com/GZHU-DVL/FDEA.

Emotional responses to color stimuli are referred to as color emotions, and relationships between color attributes and them have been discussed in past studies. In the current study, we reveal relationships between color rendering properties and color emotions pertain to face colors by using of a light system controlling color rendering property. The system we developed in the previous study has four color (red, green, blue, and yellow) LEDs to enrich color rendering property. This study provides a comprehensive discussion of color rendering properties and color preference to average face by summarizing past two studies. The first study examined color emotions against skin color under various color rendering properties. As a result, we confirmed that the color rendering property certainly affects color emotions, and we obtained regression models to represent color emotions by color rendering properties. The second study assessed relationships between color emotions against averaged face irradiated by various illuminants with different color rendering properties. As a result, we confirmed that the color rendering property certainly affects color preference, and we obtained regression models to represent color emotions by color rendering properties. Based on the model of the relationship between color emotions and color rendering properties obtained through these two studies, a comparison of skin color itself and color emotions as a face color is discussed. Specifically, the results suggest that color shifts in the red and green directions have different effects on the impressions of single colors and average faces.

This paper seeks to study the limits of the definitions of algorithmic fairness in relation to a protected variable, namely skin color. Discrimination towards the black community have existed for a long time. ML algorithms have only amplified or revealed existing discrimination. AI is a mirror that reflects the reality of our societies. The lack of a universal definition of algorithmic fairness makes it difficult to detect cases of discrimination in machine learning algorithms. We believe that independent or sensitive variables such as skin color are benchmarks that could be used to decide whether or not a decision is fair. We also recommend avoiding the use of proxy data.

We propose a method to fast capture faces in video images, in order to make full use of the screening effect of color features, in order to find the face frame in the video image more quickly, in order to cut out the face in the picture containing the face information frame, filter out the background part, and in order to expand the facial recognition algorithm application method. The method is to use the color features of each frame of the video in the video image to screen the skin color. After screening, the image is sampled and counted. If the threshold is exceeded, it is judged whether it is a human face. When determining, the face is captured and output and the threshold is corrected according to the recognized face information; if the threshold is not exceeded, skip some frames and then prejudge the frames. Experiments on videos containing face information with more complex backgrounds prove that this method can quickly find frames containing face information, and can intercept faces from frames containing face information, saving some time, and it basically does not affect the accuracy of face capture.

We present a novel high-fidelity real-time method to replace the face in a target video clip by the face from a single source portrait image. Specifically, we first reconstruct the illumination, albedo, camera parameters, and wrinkle-level geometric details from both the source image and the target video. Then, the albedo of the source face is modified by a novel harmonization method to match the target face. Finally, the source face is re-rendered and blended into the target video using the lighting and camera parameters from the target video. Our method runs fully automatically and at real-time rate on any target face captured by cameras or from legacy video. More importantly, unlike existing deep learning based methods, our method does not need to pre-train any models, i.e., pre-collecting a large image/video dataset of the source or target face for model training is not needed. We demonstrate that a high level of video-realism can be achieved by our method on a variety of human faces with different identities, ethnicities, skin colors, and expressions.

Covering face with accessories such as mask, scarf and sunglass is a common criminal activity in automated teller machine (ATM) robbery. Therefore, detection of covered face using ATM surveillance camera can be an effective solution to reduce robbery or crime. This paper presents a novel method to detect covered face from ATM surveillance camera images. Specifically, three facial features, i.e., skin color, elliptical face shape and facial width-to-height ratio (fWHR), incorporated with geometrical property of ellipse have been employed to estimate the covered region. In addition, three parameters, i.e., facial area, fWHR and covered area percentage, have been utilized for reliable classification. Experiment results demonstrate that the method can detect full covered, uncovered and partially covered faces at a correct detection rate of 98.3%, 93.3% and 97.78%, respectively. The overall correct detection rate is 96.48%, which is found to be better than previous studies. Also, the proposed method can handle faces covered with few new face hiding objects such as hijab, niqab and robber’s ski mask. Furthermore, processing time of the proposed algorithm is significantly improved while it is compared to the existing methods. The detection time varies between 31 and 67 ms which is equivalent to 15–32 frames per second.

Summary: Skin color matching in vascularized composite allotransplantation (VCA) is an important determinant of aesthetic outcomes. The process of color matching is infrequently described in the literature. The Pantone SkinTone Guide (PSTG) is a handheld tool comprised of realistic skin tone swatches with a corresponding virtual swatch system. A color match acceptability threshold (AT) is defined as the point beyond which >50% of observers deem a given skin tone pairing as unacceptable. In this study, color match acceptability thresholds were developed using the PSTG to help standardize donor-recipient color matching. Four representative colors were chosen across the skin tone spectrum. These standard colors were used to develop a survey asking participants to determine the acceptability of color pairings. Using survey results, ATs were determined for changes in lightness, undertone, and lightness and undertone combined for each of the standard colors. Inter- and intra-rater reliability were determined using Fleiss’s Kappa. Participants were more critical of skin tone pairings on the darker versus the lighter end of the spectrum, as evidenced by higher thresholds observed for lighter sample pairs. Additionally, observers were more critical of differences in skin lightness compared with differences in undertone. Intra-rater reliability was fair to substantial, and inter-rater reliability was fair to moderate. The PSTG can be used as a clinical tool to improve the aesthetic outcomes of skin-containing vascularized composite allotransplantation procedures by optimizing donor-recipient skin color matching. This can allow clinicians to complement visual judgment with quantitative reference.

Aiming at the problems of purchasing cosmetics, female always rely on their intuition or follow the fashion trend for color selection. They never understand how to select the right color to match with them. A systematic method for detecting the facial skin colors of female for the application for make-up is proposed in this study. In this method, the female face recognition with considerable images was first discussed. Next, the innovative method of capturing color with points was discussed along with making the complexion-aggregated analysis. In the study, RGB was conversed to be Lab space for collected data. The data were analyzed based on fuzzy clustering to get grouping. The results were presented in the final grouping Lab value and RGB index. Besides, a FaceRGB program was constructed to capture color automatically from 13000 images. 17 clustering results were obtained by the ColorFCM program constructed in this study. The results can be used to select the customized color of cosmetics to satisfy female and an expert system can further be made in the future.

The purpose of this study was to develop a method to create color-accurate three-dimensional (3D) printed models of the human skin, based on digital 3D scans of the skin surface. In future, this methodology could be applied to create more intricate and life-like training models for educating health-care workers on recognition of early stage pressure ulcers in patients with diverse skin tones. In this study, a transfer function was derived to compensate for color errors introduced via the 3D printing process, for a specific model of color 3D printer. The effectiveness of the transfer function was tested by measuring color errors for a set of 3D printed color blocks fabricated with and without the transfer function color correction method. Lastly, a feasibility study was conducted to demonstrate the application of the color correction algorithm to 3D printed skin models generated from 3D color scans of healthy human subjects of varying skin tones.