Skin Color Model Research Articles

Hand Gesture Recognition System Based on Indian Sign Language Using SVM and CNN

Hand gestures (HGs) have been extensively acknowledged as a communication standard for the deaf and dumb school system in the Indian sign language (ISL). We suggested an automated Indian sign language recognition (ISLR) method for English in this study. Initially, the hand region is segmented using the Grasshopper optimization algorithm (GOA) based on the skin color model method. The segmentation’s efficacy is further assessed using three techniques: GOA-based skin color detection algorithm (SCDA), particle swarm optimization-based SCDA (PSO-SCDA), and artificial bee colony-based SCDA (ABC-SCDA). Finally, a database for the retrieved motions representing distinct English alphabets is created. For gesture recognition, the system is trained, and a template-based matching approach is used. The classification technique employs the support vector machine (SVM) and the convolution neural network (CNN). Our suggested recognition method achieved the highest accuracy of 97.85% with GOA-SCDA, compared to 89.29% and 93.96% with PSO-SCDA and ABC-SCDA, respectively. Furthermore, CNN outperformed SVM in classification, with a high accuracy of 99.2% and precision of 81.8%.

Skill assessment based on clutch use in cross-platform robot-assisted surgery.

Many studies have investigated the transfer of skills between laparoscopic and robot-assisted surgery (RAS). These studies have considered time, error, and clinical outcomes in the assessment of skill transfer. However, little is known about the specific operations of the surgeon. Clutch control use is an important skill in RAS. Therefore, the present study aimed to propose a novel objective algorithm based on computer vision that can automatically evaluate a surgeon's clutch use. Additionally, the study aimed to evaluate the correlation between clutch metrics and surgical skill on different surgical robot platforms. The robotic surgery training center of Wuhan University trained 30 laparoscopic surgeons as the study group between 2023 and 2024. Laparoscopic surgeons were trained by combining robotic simulator exercises and RAS animal experiments. During the training, video and hand movement data were collected. Hand movements identified by a skin-color model were combined with labeling information to classify clutch use. The metrics were validated on different robotic platforms (dv-Trainer, EDGE MP1000, Toumai™ MT1000, and DaVinci Xi system) and among surgeons with different surgical skill levels. On the robotic simulator, clutch accuracy in the expert group was significantly higher than in the study group for all tasks. No significant differences were observed in the number of clutches between the expert and study groups. In the RAS experiment, the number of clutches decreased significantly for both study and expert groups. The accuracy was maintained at a high level in the expert group but decreased rapidly in the study group. We proposed a new objective assessment of surgical skills, clutch use metrics, in cross-platform RAS. Additionally, we verified that the metrics significantly correlated with the surgical skill levels of the surgeons.

Digital Apparel Try-Ons: A Technological Odyssey into the World of Virtual Dressing Rooms and Consumer Engagement

This individual can now enjoy internet shopping whenever they want and from anywhere in the world thanks to the invention of smartphones and tablets. Online shopping has unquestionably taken the place of conventional methods for purchasing apparel and daily necessities. People benefit from a reputation as they pick internet buying. Almost all online retailers offer payment on delivery, free shipment, and special rates these days. These internet merchants have eliminated the hassles of parking, stuck in traffic, and standing in line to make purchases. They have also helped folks who consistently lament their lack of time. This is the primary justification for why most consumers now shop online. They may easily access a wide choice of affordable prices here, and they also get quick customer service and free home delivery. There is no denying that these are a few of the alluring aspects that draw customers in. Even though there is a minor problem that can make people lose interest in online buying, they might not be able to try on garments in such situations. By establishing a virtual changing room setting, the aim is to improve accessibility and time efficiency for trying on clothing. Our proposed methodology consists of three main components: skin color recognition, model alignment, and user image extraction from the video stream. Through user extraction, we may create an augmented reality environment by taking the user area out of the video stream and superimposing it on top of a simulated world within the user interface. To match the 2D cloth models with the use of scaling, rotation, and positioning using the 3D locations of the joints.

A Hybrid Photorealistic Architecture Based on Generating Facial Features and Body Reshaping for Virtual Try-on Applications


 
 
 Online shopping using virtual try-on technology is becoming popular and widely used for digital transformation because of sustainably sourced materials and enhancing customers’ experience. For practical applicability, the process is required for two main factors: (1) accuracy and reliability, and (2) the processing time. To meet the above requirements, we propose a state-of-the-art technique for generating a user’s visualization of model costumes using only a single user portrait and basic anthropometrics. To start, this research would summarize different methods of most virtual try-on clothes approaches, including (1) Interactive simulation between the 3D models, and (2) 2D Photorealistic Generation. In spite of successfully creating the visualization and feasibility, these approaches have to face issues of their efficiency and performance. Furthermore, the complexity of input requirements and the users’ experiments are leading to difficulties in practical application and future scalability. In this regard, our study combines (1) a head-swapping technique using a face alignment model for determining, segmenting, and swapping heads with only a pair of a source and a target image as inputs (2) a photorealistic body reshape pipeline for direct resizing user visualization, and (3) an adaptive skin color models for changing user’s skin, which ensures remaining the face structure and natural. The proposed technique was evaluated quantitatively and qualitatively using three types of datasets which include: (1) VoxCeleb2, (2) Datasets from Viettel collection, and (3) Users Testing to demonstrate its feasibility and efficiency when used in real-world applications
 
 

Object Detection in the Image using Matlab

In videoconferencing, the image quality is significantly affected by the illumination condition.Unsatisfactory illumination conditions may lead to underexposure or overexposure of the area ofinterest, in particular a human face. To resolve this issue, we propose a solution to automaticallyimprove image quality by correcting exposure and enhancing contrast. Our work is characterized by amethod for automatically building a skin-color model and a novel contrast enhancement approach.Some techniqu an reduce the computational cost are also introduced. Experimental results show thatobvious improvement in image quality is achieved while the computation overhead is very small. Theproposed solution can be integrated into videoconferencing systems and is especially suitable forscenarios where low-complexity computing is required. From our experiments, it is found that there isan optimal resolution to balance between speed and accuracy, while we show in our experiments thatthe resolution has apparent influence on both the accuracy and the computing time.

Lightweight real-time hand segmentation leveraging MediaPipe landmark detection

Real-time hand segmentation is a key process in applications that require human–computer interaction, such as gesture recognition or augmented reality systems. However, the infinite shapes and orientations that hands can adopt, their variability in skin pigmentation and the self-occlusions that continuously appear in images make hand segmentation a truly complex problem, especially with uncontrolled lighting conditions and backgrounds. The development of robust, real-time hand segmentation algorithms is essential to achieve immersive augmented reality and mixed reality experiences by correctly interpreting collisions and occlusions. In this paper, we present a simple but powerful algorithm based on the MediaPipe Hands solution, a highly optimized neural network. The algorithm processes the landmarks provided by MediaPipe using morphological and logical operators to obtain the masks that allow dynamic updating of the skin color model. Different experiments were carried out comparing the influence of the color space on skin segmentation, with the CIELab color space chosen as the best option. An average intersection over union of 0.869 was achieved on the demanding Ego2Hands dataset running at 90 frames per second on a conventional computer without any hardware acceleration. Finally, the proposed segmentation procedure was implemented in an augmented reality application to add hand occlusion for improved user immersion. An open-source implementation of the algorithm is publicly available at https://github.com/itap-robotica-medica/lightweight-hand-segmentation.

Background region based Face orientation prediction through HSV skin color model and K-Means clustering

Background region based Face orientation prediction through HSV skin color model and K-Means clustering

Study on A face detection method based on elliptic skin color model

Human face is a natural structural target with abundant details, and its detection results are easily affected by facial details, expressions and posture changes. In color images, the distribution of skin color is not affected by the changes of facial details, expression and posture, and the speed of skin color detection is very fast. This paper proposes a face detection algorithm based on elliptic skin color model. In YCbCr color space, the elliptic skin color model and logistic regression analysis were used to determine the skin color probability of each point, and the pixels of each point were mapped to [0, 1]. Based on Ostu method, a parallel genetic algorithm was used to determine the threshold of skin color segmentation to segment the face region. The results show that this method improves the speed of face detection and has good robustness to posture and expression changes.

Kinect and Few-Shot Technology-Based Simulation of Physical Fitness and Health Training Model for Basketball Players in Plateau Area.

Players in modern basketball have a lot of physical contact, a lot of bumps, and a lot of physical struggles. The competition for the ball, whether in the air or on the ground, is fierce, putting higher demands on the players' physical abilities. Coaches frequently use plateau physical training, which is very effective in developing athletes' cardiopulmonary function, among many other training methods. The proportional length and active area of arms are obtained using the skin color model of the human body, the angle and posture information of each joint is extracted from dynamics, and the 3D posture of arms and dynamic arms is trained and recognized in this paper, which is based on Kinect. The findings revealed that mild hypoxia in the plateau significantly lowered basketball players' performance and that basketball players' maximum heart rate and 1-minute heart rate recovery in high-intensity exercise were lower than those in flat area training.

Gesture Recognition Analysis Based On Monocular Vision Relying On Human-Computer Interaction Filtering Algorithm

The progress of vision-based gesture recognition technology has introduced a brand new way of human-computer interaction (HCI). However, due to the complexity of feature extraction, gesture recognition based on monocular vision has become a research hotspot and bottleneck. In this paper, static gestures are combined with the CSS feature description to extract the skin-color model based on the HCI filtering algorithm first. Subsequently, the fusion feature is established. Finally, feature recognition is performed based on the HCI filtering algorithm to implement gesture recognition analysis based on monocular vision. The practice has demonstrated that the proposed algorithm has a relatively high recognition rate for complex gestures with the sign language of high local similarity.

Analysis of the Current Situation of Piano Education in Colleges and Universities in the Information Age and Research on Countermeasures

In view of the low application ability of piano improvisational accompaniment of music majors, this paper proposes a method of big data combined with MIDI keyboard and Kinect depth sensor to achieve the purpose of recognizing chord progression and judging fingering when students perform, and realizes the auxiliary teaching system. Firstly, the information of color and depth images is obtained, and the state transition diagram of chord transposition and chord gesture template library are constructed as the system initialization conditions. Secondly, using the traditional skin color modeling and background difference method as well as the current depth data, the gesture recognition is realized by template matching. Finally, the correctness of chord progression is judged, and comprehensive fingering application is used to score and evaluate. The experimental results show that the system has high robustness and can be effectively applied to piano teaching.

Study on face detection based on an improved Gaussian skin color model

An improved Gaussian skin color model is proposed for face detection applications. It overcomes the issue that the traditional models become ineffective in extracting the skin color region in the presence of diverse skin colors and illuminations. It is a two-dimensional gaussian mixture model based on H-SV (i.e., an improved HSV) and YCbCr color spaces. Experimental results from several face image databases show that the detection accuracy is improved by 35.8% and 6.3%, respectively, compared to a traditional gaussian skin model and Adaboost algorithm.

Real-Time Hand Gesture Recognition

With the rapid development of computer vision, the demand for interaction between humans and machines is becoming more and more extensive. Since hand gestures can express enriched information, hand gesture recognition is widely used in robot control, intelligent furniture, and other aspects. The paper realizes the segmentation of hand gestures by establishing the skin color model and AdaBoost classifier based on haar according to the particularity of skin color for hand gestures and the denaturation of hand gestures with one frame of video being cut for analysis. In this regard, the human hand is segmented from a complicated background. The camshaft algorithm also realizes real-time hand gesture tracking. Then, the area of hand gestures detected in real-time is recognized by a convolutional neural network to discover the recognition of 10 common digits. Experiments show 98.3% accuracy.

Automatic Ear Localization Using Entropy-Based Binary Jaya Algorithm and Weighted Hausdorff Distance

This paper presents an automatic human ear localization technique for handling uncontrolled scenarios such as illumination variation, poor contrast, partial occlusion, pose variation, ear ornaments, and background noise. The authors developed entropy-based binary Jaya algorithm (EBJA) and weighted doubly modified Hausdorff distance (W-MHD) to use edge information rather than pixels intensity values of the side face image. First, it embodies skin segmentation procedure using skin color model and successively remove spurious and non-ear edges which reduces the search space of the skin regions. Secondly, EBJA is proposed to trace dense edge regions as probable ear candidates. Thirdly, this paper developed an edge based weight function to represent the ear shape along with for the edge based template matching using W-MHD to identify true ear from a set of probable ear candidates. Experimental results using publicly available benchmark datasets demonstrate the competitiveness of the proposed technique in comparison to the state-of-the-art methods.

A Face Detection Method Based on Skin Color Features and AdaBoost Algorithm

The face detection based on skin color features is greatly affected by illumination and the face detection based on AdaBoost algorithm is greatly affected by multi-pose of face, and this means that the large face deflection will make the detection less accurate. So we investigated a face detection method that combines the skin color model and the improved AdaBoost algorithm in this paper. This method first judges whether the collected face image needs illumination compensation. The image is converted from RGB color space to YCbCr color space with higher pixel clustering characteristics and distribution rules by whether it is needed or not. Meanwhile, the candidate regions containing face images are carried out through morphological process and shape select after establishing the Gaussian model. Finally, the face detection results are obtained through detecting the candidate regions with face images by improved AdaBoost algorithm. A large number of experimental results show that this new method can effectively reduce the false detection rate and improve the efficiency of face detection.

A Face Detection Method Based on Skin Color Model and Improved AdaBoost Algorithm

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.

Efficient covering of target areas using a location prediction-based algorithm

Due to the rapid development of the high-speed wired and wireless Internet, image contents including objects with exposed personal information are being distributed freely, which is a social problem. In this paper, we introduce a method of robustly detecting a target object with facial region exposed from an image that is quickly entered using skin color and a deep learning algorithm and effectively covering the detected target object through prediction. The proposed method in this paper accurately detects the target object containing facial region exposed from the image entered by applying an image adaptive skin color model and a CNN-based deep learning algorithm. Subsequently, the location prediction algorithm is used to quickly track the detected object. A mosaic is overlapped over the target object area to effectively protect the object area where the facial region is exposed. The experimental results show that the proposed approach accurately detects the target object including the facial region exposed from the continuously entered video, and efficiently covers the detected object through mosaic processing while quickly tracking it using a prediction-based tracking algorithm. The tracking-based target covering method proposed in this study is expected to be useful in various practical applications related to pattern recognition and image security, such as content-based image retrieval, real-time surveillance, human–computer interaction, and face detection.

Research of Face Detection Based on Skin Color Feature

Face image contains many pattern features and this paper studies a face detection algorithm based on skin color features. Through the analysis of the face detection method based on skin color features, the calculation process and principle are understood, and the algorithm and detection techniques such as color space, preprocessing face image detection technology. Building a corresponding skin color model based on face characteristics are analyzed in depth. Because the skin color only occupies a specific range in the primary color of color space, the probability likelihood graph of skin color is obtained by calculating skin color similarity. Then the binary image of the possible face region is obtained by smooth filtering, threshold segmentation, and other operations. Then the area center and height ratio are used to calculate the center of mass and deflection angle, and the correlation coefficient is matched with the face module. Finally, the face area is detected and marked. Test results show that the algorithm can achieve face detection well.

A Survey on Artificial Intelligence in Chinese Sign Language Recognition

Chinese Sign Language (CSL) offers the main means of communication for the hearing impaired in China. Sign Language Recognition (SLR) can shorten the distance between the hearing-impaired and healthy people and help them integrate into the society. Therefore, SLR has become the focus of sign language application research. Over the years, the continuous development of new technologies provides a source and motivation for SLR. This paper aims to cover the most recent approaches in Chinese Sign Language Recognition (CSLR). With a thorough review of superior methods from 2000 to 2019 in CSLR researches, various techniques and algorithms such as scale-invariant feature transform, histogram of oriented gradients, wavelet entropy, Hu moment invariant, Fourier descriptor, gray-level co-occurrence matrix, dynamic time warping, principal component analysis, autoencoder, hidden Markov model (HMM), support vector machine (SVM), random forest, skin color modeling method, k-NN, artificial neural network, convolutional neural network (CNN), and transfer learning are discussed in detail, which are based on several major stages, that is, data acquisition, preprocessing, feature extraction, and classification. CSLR was summarized from some aspect as follows: methods of classification and feature extraction, accuracy/performance evaluation, and sample size/datasets. The advantages and limitations of different CSLR approaches were compared. It was found that data acquisition is mainly through Kinect and camera, and the feature extraction focuses on hand’s shape and spatiotemporal factors, but ignoring facial expressions. HMM and SVM are used most in the classification. CNN is becoming more and more popular, and a deep neural network-based recognition approach will be the future trend. However, due to the complexity of the contemporary Chinese language, CSLR generally has a lower accuracy than other SLR. It is necessary to establish an appropriate dataset to conduct comparable experiments. The issue of decreasing accuracy as the dataset increases needs to resolve. Overall, our study is hoped to give a comprehensive presentation for those people who are interested in CSLR and SLR and to further contribute to the future research.

Ethnicity Classification Based On Facial Features Using Viola-Jones Algorithm

Biometric recognition system can use race classification to identify the human globally with a particular identity. This paper proposes Support Vector Machine and will compare the result with K-Nearest Neighbor for classification of people into two major races namely Indonesian western and eastern races. Firstly, the proposed classification method extracts the distinct primary facial feature and skin color model of the given face with Viola-Jones Algorithm to effectively classify the races. To increase the accuracy, the sample must not contain any background of other people skin, no movement and the pictures were taken from the mobile camera with no beauty filter.