No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
A robust skin color based face detection algorithm
Abstract
In this paper, a detailed experimental study of face detection algorithms based on "Skin Color" has been made. Three color spaces, RGB, YCbCr and HSI are of main concern. We have compared the algorithms based on these color spaces and have combined them to get a new skin color based face detection algorithm which gives higher accuracy. Experimental results show that the proposed algorithm is good enough to localize a human face in an image with an accuracy of 95.18%.
... The RGB color space [13] consists of the three additive primaries: red, green and blue. Spectral components of these colors combine additively to produce a resultant color. ...
... YCbCr color space [13] has been defined in response to increasing demands for digital algorithms in handling video information, and has since become a widely used model in a digital video. It belongs to the family of television transmission color spaces. ...
... Since hue, saturation and intensity are three properties used to describe color, it seems logical that there be a corresponding color model, HSI [13]. When using the HSI color space, you don't need to know what percentage of blue or green is required to produce a color. ...
The segmentation of objects whose color-composition is not trivial represents a difficult task. In this work we propose a fuzzy algorithm application for the segmentation of such objects. It is chosen; by the characteristics that it represents the face segmentation. A priori knowledge about spectral information for certain face skin region classes is used in order to classify image in fuzzy logic classification procedure. The basic idea was to perform the classification procedure first in the supervised and then in fuzzy logic manner. Some information, needed for membership function definition, was taken from supervised maximum likelihood classification. The system uses three membership functions which are taken as Gaussian distribution curve. For real time needs, the system is implemented on an FPGA.
... 4. The eigenvector and eigenvalue are calculated using the estimated covariance matrix. For the N-dimensional vector, there will be N eigenvalues and eigenvectors (Singh et al., 2003). 5. Finally, the eigenvalue is sorted out from high to low, then the first N eigenvectors that have large variances are chosen and removed the ones with low variance, so that could reduce the dimensionality. ...
... The location and size of each face image remain similar (Singh et al., 2003). ...
... The size and location of each face image must remain similar (Singh et al., 2003). ...
Face recognition is a biometric technique that can be used for a variety of purposes, such as national security, access control, identity fraud, banking, and finding missing children. Faces are highly dynamic and facial features are not always easily extracted, which can lead to discarding textural information like the smoothness of faces, a hairstyle that, might contain strong identity information. In addition, brightness, scale, and facial expressions play a significant role in the face-recognizing process. Therefore, face recognition is considered as a difficult problem. To figure out this problem effective methods using databases techniques are needed. This paper describes face recognition methods and their structure. Based on Wen Yi Zhao and Rama Chellappa work the face recognition methods are divided into three groups: a holistic approach, feature-based approach, and hybrid approach, where Principal Component Analysis PCA, a holistic approach method, is presented as a mathematical technique that can assist the process of face recognition. Also, the paper shows how the PCA is used to extract facial features by removing the principal components of the available multidimensional data.
... One way to increase tolerance towards changes in image density is converting an RGB image to the color space that the intensity and color of a separate and use only a part colorimetric detection. In this paper, we present a suitable face detection algorithm that can detect faces with different depth and multi faces in both indoor and outdoor environments [2]. ...
... The red, green and blue color components are highly correlated. This makes it difficult to execute some image processing algorithms [2]. ...
... It belongs to the family of television transmission color spaces. The family includes others such as YUV and YIQ [2]. ...
The goal of face detection is to locate all regions that contain a face. This paper has a simple face detection procedure, first to segment skin region from an image, and second, to decide these regions contain human face or not. Our procedure is based on hybrid skin color segmentation using three color spaces RGB, YC b C r and HIS and human face features using entropy. For the purpose of extracting feature, rather than looking at the whole image of the face, and put the entropy based on the selection of skin region, which selects high informative segments of the facial image, compared with entropy of ORL image using the Euclidean distance .Also the golden ratio and the size of skin region decide where this region is face or no through the fuzzy system. Fuzzy logic got great acceptance of the various fields therefore it used in this paper to cover the difference in the parameter of face. The method provides a suitable method for extracting information. The proposed method has been tested on various real images and its performance is found to be quite satisfactory with detection accuracy 94.74 %.
... Detection can be used as a preliminary step in some computer vision applications such as nudity recognition on websites, face detection, skin disease detection, etc. Thus, in the field of skin color detection several works have been carried out through different approaches among which we can cite those of Singh et al. (2003) who used skin color detection for the localization of faces on images. For this purpose, they combined the three color spaces RGB, YCbCr and HSI to obtain a new face detection algorithm based on skin color, the experimental results of which show an accuracy of 95.18%. ...
... Furthermore, in our system we use the components of the RGB color space to authenticate an individual. We could have better results by using a combination of RGB, YCbCr and HSI color spaces as proposed by Singh et al. (2003). On another level, it is worth noting the significant effect of the factors causing intra-class variations. ...
This manuscript presents the design of a new approach of human skin color authentication. Skin color is one of the most popular soft biometric modalities. Since a soft biometric modality alone cannot reliably authenticate an individual, this new system is designed to combine skin color results with other pure biometric modalities to increase recognition performance. In the classification process, we first perform facial skin detection by segmentation using the thresholding method in the HSV color space. Then, the K-means algorithm of the clustering method is used to determine the dominant colors on the skin pixels in the RGB model. Variations according to the R, G and B components are recorded in a reference model to enable an individual’s identity to be predicted on the basis of 30 clusters. Experimental results are promising and give a false acceptance rate (FAR) of 29.47% and a false rejection rate (FRR) of 70.53%.
... Likewise, the color green and pure red can be depicted as (0, 255, 0) and (0, 0, 255), respectively. [7,8] In RGB color model, a specific color could be made by different combinations of each color component. Thus, in some specific application where the purpose is to choose a specific range of color spectrum, it will be difficult to use the image processing. ...
... Thus, in some specific application where the purpose is to choose a specific range of color spectrum, it will be difficult to use the image processing. [8] ...
Background:
The objective of the study is to evaluate elastography ultrasound findings in patients with scleroderma (SS) and to clarify the effectiveness of elastosonography to differentiate scleroderma lesions from any skin lesion considering tissue elasticity.
Methods:
Thirty-six SS patients definite diagnosis of systemic sclerosis according to American College of Rheumatology criteria and 36 healthy subjects were enrolled. Volar aspect of the middle forearm and arm in addition to the dorsal aspect of the fingers were evaluated by sonoelastography. The RGB (red, green, blue) image is a three-dimensional matrix. A color image RGB is an M × N × 3 array of color pixels. The total pixels, total blue pixels, and blue index compared between SS cases and controls.
Results:
Mean age of patients was 41.3 ± 10.3 years and mean age of controls was 39.8 ± 9.3 years. Mean-modified Rodnan skin score of the whole body was 11.9 and mean duration of disease was 6.2 years. Mean total blue pixels in the arm were significantly different between cases and controls. Mean total image pixels, total blue pixels, and blue index in the forearm were significantly different between cases and controls. Elastography findings in the finger were not significantly different between cases and controls.
Conclusions:
Sonoelastography could be used for evaluating skin of forearm in sclerodermic cases which will be helpful for disease evaluation in clinical course.
... Two color coordinate systems: the Red, Green and Blue (RGB), and the Hue, Saturation and Intensity (HSI) have been chosen for testing the proposed detection algorithm in addition to the gray-scale system. The RGB color model is an additive model, in which red, green, and blue light are added together in various ways to reproduce target colors [40]. The HSI color model has three components: hue, saturation, and intensity, that are used to describe a certain color [9]. ...
... All these image representation systems are sensitive to illumination changes. The RGB is more sensitive to illumination intensity [40], and the B component is the most sensitive one. The structuremeasure (S-measure) [15] and Enhanced-alignment measure (E-measure) [16] can be used to evaluate the structure similarity between various image channels (R, B, and G). ...
Digital image forgery detection is an important task in digital life as the image may be easily manipulated. This paper presents a novel blind tampering detection algorithm for images acquired from digital cameras and scanners. The algorithm is based on applying homomorphic image processing on each suspicious image to separate illumination from reflectance components. In natural images, it is known that the illumination component is approximately constant, while changes can be detected in tampered ones. Support Vector Machine (SVM) and Neural Network (NN) classifiers are used for classification of tampered images based on the illumination component, and their results are compared to obtain the best classifier performance. The Receiver Operating Characteristic (ROC) curve is used to depict the classifier performance. Three different color coordinate systems are tested with the proposed algorithm, and their results are compared to obtain the highest accuracy level. Joint Photographic Experts Group (JPEG) compressed images with different Quality Factors (QFs) are also tested with the proposed algorithm, and the performance of the proposed algorithm in the presence of noise is studied. The performance of the SVM classifier is better than that of the NN classifier as it is more accurate and faster. A 96.93% detection accuracy has been obtained regardless of the acquisition device.
... Under certain lighting conditions, color is orientation invariant [11]. Good skin color pixel classification should provide coverage of all different skin types (blackish, yellowish, brownish, whitish, etc.) and cater for as many different lighting conditions as possible [12]. ...
... 11) and divide by 2 to determine the center of face as shown inFigure (12) which has the face skin location (skin image). Then by returning to the original image use the skin image to get face image, Figure (13).ii. ...
... Then eyes, ears, nose facial features can be extracted from the binary image by considering the threshold for areas which are darker in the mouth than a given threshold. After getting the triangle, it is easy to get the coordinates of the four corner points that form the potential facial region [27] [28]. ...
... For those pixels that satisfy (3), they undergo a conversion to the {Y, I, Q} space in order to extract the chroma phase angle histogram statistics. Upon doing so, a histogram of the phase angle arctan(Q/I) is obtained in order to extract the statistics of the proportion of pixels falling outside of the bounds as described by (2). In the following examples, a set of images for each selected example has been provided, where in: - ...
span lang="EN-US">Generally, chroma phase or hue offset issues within a scene are hard to detect, without a reference or context (i.e. some apriori knowledge about how certain objects within the scene should actually appear in terms of their hue). Moreover, when it comes to skin/flesh tones, hue deviation can be noticeable and can markedly degrade the viewer quality of experience (QoE) , whenever it does occur. However a lot of research has gone into flesh tone detection, specifically, the color gamut within which flesh tone is present. This topic has been well documented in the literature with respect to various color spaces: red, green, blue (RGB) and YIQ. Therefore, overall issues with chroma offset or hue within the video content could potentially be approached by extracting and analyzing a reliable reference, such as skin or flesh tone (if present), within some allowable deviation. This involves machine learning (ML) based facial recognition and tracking followed by skin tone region recognition within the detected facial sequence (i.e. Region of Interest). The skin region serves as a ‘self-reference’ in order to discern any inherent phase offset within the content. Finally, the angular chroma deviation discerned can then be used for subsequent correction as well.
</p
... There are many challenges in detection of the face like presence of glasses, beard, mustaches, face posture, expression of face, face occlusion, etc. Also the shape and size of the face and skin color varies from person to person [4]. ...
A security system based on biometrics of a person depends on the features of a person. These can be physical features or behavioral features. User authentication is necessary for security control system and intelligent human computer interface, so that only the real user accesses it. We present a method for the authentication of person with help of multimodal data which includes face and speech. Agent based system has been used for fusion of these data which gives us a more reliable, efficient and secure biometric system. A method for face detection for recognition of faces has been given. For speech recognition Artificial Neural Network has been used. Both these combined together identify the speaker. This multimodal biometric method will give highly reliable secure system.
... As a result of the high connection between's color portions: blue , green, and red as each fragment is obligated with the influence of luminance of the light power of nature, so gets frustration concerning many picture getting ready applications. In utilitarian, this model isn't fitting to portray colors similarly as human comprehension [13]. ...
During last 10 years people are very much attracted to face recognition systems and they are very much eager to solve the issues related to face recognition system. It helped them very much in the field of electronics and uses over pattern unlocking and password entering system. There are numerous applications as for security, affectability and mystery. Detection of a face is the most significant and initial step of recognition framework. This article demonstrates a new method to face recognition system using color and template of an image. Whatever the background it may go to be, our system will detect the face, which is an important stage for face detection. The pictures utilized in this framework for Face detection are the color images, while the images used for the Face Recognition are the Gray images which are converted from color pictures. The illumination compensation technique is applied on all the images for removing the effect of light. The Red, Green, and Blue values of each pixel will be converted to YCbCr space. Based on the probability of each pixel in terms of Cb, Cr values, we extract the skin pixels from the query image,. The positive probability shows a “skin pixel”, while the negative probability shows “not a skin pixel”. Finally the face is projected. In face recognition, we used 4 templates of different sizes for Gabor image content extraction. Finally we employed the relevance feedback mechanism to retrieve the most similar images. If the user did not satisfy with the given results he can give the correct images to the system from the displayed images. Exploratory outcomes demonstrate that the demonstrated system is adequate to recognize face of a human face in a picture with an exactness of 94%.
... Непросто встановити зображення кольору шкіри людини. Однак є алгоритми виявлення обличчя на основі кольору шкіри (Singh, et al., 2003). ...
The purpose of the article is to research, analyze and consider the general problems and prospects of using existing approaches to face recognition (areas of application, features and differences). The research methodology consists of semantic analysis methods of the basic concepts in this subject area (theory and practice of pattern recognition, in particular, facial images). The article considers the existing approaches to the development of systems for face recognition. The novelty of the research is the solution of facial recognition problems to determine access rights and authentication. Conclusions. The existing problems analyzed and the prospects for using facial recognition algorithms are becoming more accurate. Facial recognition has become an important part of artificial intelligence because it is used in social media, digital cameras and smart home automation.
... The lighting affects a lot in the color images to find the face, so they found the property of chrome, which adds good properties in the image [9]. They are tried to build face detection algorithm through face or skin color [10]. The stick contain sensor that provide distance between blind people and the objects by use Ultrasonic sensor, infrared (IR), Light dependent resistor (LDR) [11,12,13,14]. ...
Physical disability has affected many people’s lives across the world. One of these disabilities that strongly affected some large category of people is visual lose. Blind people often face difficulties in moving around freely such as: in crossing the street, in reading, driving or socializing. They often rely on using certain aid devices to reach certain places or perform any other daily activities such as walking sticks. There are ongoing scientific researches in the area of rectifying blindness, but it has to go long way to achieve the solution. Also, there are research unleashes the ideas of assisting the blind people deficiency but lacks in technological aspects of implementation. This research project aims at helping blind people of all categories to achieve their day to day tasks easier through the use of a smart device. By using artificial intelligent and image processing, this smart device is able to detect faces, colors and deferent objects. The detection process is manifested by notifying the visually impaired person through either a sound alert or vibration. Additionally, this study presents a palpable survey that entails visually impaired people from the local community. Subsequently, the project uses both Open CV and Python for programming and implementation. The exertion of this project prototype investigates the algorithms which are used for detecting the objects. Also, it demonstrates how this smart device could detects certain physical object and how it could send a warning signal when faced by any obstacles. Overall, this research will be a positive addition in the world of health care sector by supporting blind people with the use of smart technology
... However, the emergence of abstract mathematical methods such as eigenfaces [26,27] has introduced another approach in face recognition, gradually leaving behind the anthropometric approach. However, there still some features such as skin-color that are relevant for face detection [28]. It is however, essential to present abstractions and combat the challenges from a pure computational or mathematical view point. ...
One of the most pertinent applications of image analysis is face recognition and one of the most common genetic disorders is Down syndrome (DS), which is caused by chromosome abnormalities in humans. It is currently a challenge in computer vision in the domain of DS face recognition to build an automated system that equals the human ability to recognize face as one of the symmetrical structures in the body. Consequently, the use of machine learning methods has facilitated the recognition of facial dysmorphic features associated with DS. This paper aims to present a concise review of DS face recognition using the currently published literature by following the generic face recognition pipeline (face detection, feature extraction, and classification) and to identify critical knowledge gaps and directions for future research. The technologies underlying facial analysis presented in recent studies have helped expert clinicians in general genetic disorders and DS prediction.
... As expression occurs on the face, major focus of processing should be on facial image which requires face detection or facial landmarks detection in an input image. Depending on the approach to utilize face knowledge, face detection methods can be categorized as -Feature-based approach -Image-based approach Face detection in feature-based approach is done by lowlevel features derivation while in an image-based approach, face is represented as 2D array and directly classified by trained classifier [11][12][13][14][15]. In [16], various face detection methods are categorized into approaches based on rigid templates (includes boosting-based methods such as Viola-Jones face detection algorithm and its variations or the application of deep neural networks), and non-rigid templates (includes deformable models that describe the face by its parts) Next step of automated FER is facial feature extraction and representation, where interpretation of facial variations caused by facial expressions is extracted by using geometric features or/and appearance-based features [17][18][19][20]. ...
Among various means of communication, the human face is utmost powerful. Persons suffering from Parkinson’s disease (PD) experience hypomimia which often leads to reduction in facial expression. Hypomimia affects in social interaction and has a highly undesirable impact on patient’s as well as his relative’s quality of life. To track the longitudinal progression of PD, usually Movement Disorder Society’s Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) is used in clinical studies and item 3.2 (i.e., facial expression) of MDS-UPDRS defines hypomimia levels. Assessment of facial expressions has traditionally relied on an observer-based scale which can be time-consuming. Computational analysis techniques for facial expressions can assist the clinician in decision making. Intention of such techniques is to predict objective and accurate score for facial expression. The aim of this paper is to present up-to-date review on computational analysis techniques for measurement of emotional facial expression of people with PD (PWP) along with an overview on clinical applications of automated facial expression analysis. This led us to examine a pilot experimental work for masked face detection in PD. For the same, a deep learning-based model was trained on NVIDIA GeForce 920M GPU. It was observed that deep learning-based model yields 85% accuracy on the testing images.
... Skin color Based: As expressed by author in [6], algorithms are compared using 3 colors. This mixture of color leads to face detection algorithm with color replacement. ...
Facial expression is a major area for non-verbal language in day to day life communication. As the statistical analysis shows only 7 percent of the message in communication was covered in verbal communication while 55 percent transmitted by facial expression. Emotional expression has been a research subject of physiology since Darwin’s work on emotional expression in the 19th century. According to Psychological theory the classification of human emotion is classified majorly into six emotions: happiness, fear, anger, surprise, disgust, and sadness. Facial expressions which involve the emotions and the nature of speech play a foremost role in expressing these emotions. Thereafter, researchers developed a system based on Anatomic of face named Facial Action Coding System (FACS) in 1970. Ever since the development of FACS there is a rapid progress of research in the domain of emotion recognition. This work is intended to give a thorough comparative analysis of the various techniques and methods that were applied to recognize and identify human emotions. This analysis results will help to identify the proper and suitable techniques, algorithms and the methodologies for future research directions. In this paper extensive analysis on the various recognition techniques used to identify the complexity in recognizing the facial expression is presented. This work will also help researchers and scholars to ease out the problem in choosing the techniques used in the identification of the facial expression domain.
... S1. For the color analyses, in the CIELAB color space the a* value represents the redness and greenness from positive to negative, the b* value represents the yellowness and blueness from positive to negative and the L* value represents the lightness (Singh, Chauhan, Vatsa, & Singh, 2003). Tomatoes used in this study were harvested after reaching its maturity level which was decided visually (spherical, uniform red color formation), so a* values of accessions were always positive and there were no negative values which is related with the unripe stages. ...
In this study, 50 tomato landraces grown in Turkey were investigated in terms of their secondary metabolite profiles. Each accession was planted in 2016 and 2017 in 3 replicates in an open field. In this study, color, pH and brix of the fruit samples were measured and an unbiased LCMS-based metabolomics approach was applied. Based on Principal Components Analysis (PCA) and Hierarchical Cluster Analysis (HCA) of the relative abundance levels of >250 metabolites, it could be concluded that fruit size was the most influential to the biochemical composition, rather than the geographical origin of accessions. Results indicated substantial biodiversity in various metabolites generally regarded as key to fruit quality aspects, including sugars; phenolic compounds like phenylpropanoids and flavonoids; alkaloids and glycosides of flavour-related volatile compounds. The phytochemical data provides insight into which Turkish accessions might be most promising as starting materials for the tomato processing and breeding industries.
... • Color Images -Typically, skin color is used to find faces. But the drawback would be if light conditions are weak, this does not work quite well [33]. ...
This thesis presents a hardware architecture, IoT-Edge-Server, of a diverse embedded system combining the applications of a smart city, smart building, or smart agricultural farm. First of all, we improve computation time by integrating the idea of edge computing on Raspberry Pi, CPU, and Field Programmable Gate Array (FPGA) which processes different algorithms. Second, the hardware processors are connected to a server which can manipulate the entire system and also possess storage capacity to save the system’s important data and log files. More specifically, the hardware computes data from – 1) a non-standardized Bluetooth Low Energy (BLE) Mesh System, and 2) a Security Monitoring System. The BLE Mesh System has one master and three slave devices while the Security Monitoring System has a Passive Infrared Sensor (PIR) and a webcam to detect motion. Experimental results prove that using the phenomena of edge computing also known as fog computing demonstrates an improvement in computation speed and data privacy. Although the results from the Raspberry Pi and general purpose CPU show drastic improvement, our expectations were more than that from the systems. To enhance it even further, we also propose third computing device which is a hardware accelerator. We present a synthesis of the well-known Viola-Jones face detection algorithm on Xilinx software and platform - Vivado and FPGA as Nexys 4 Artix-7 device, due to hardware accelerators' fast computation ability. Compared with the prior work on the Altera platform proposed in, our work reduces the slice count by 1018. Additionally, the power consumption of the implementation is 714 mW, including 15\% as the static cost and 85\% as the dynamic power dissipation. Furthermore, the design details of the components of the structure, such as the generation of an integral image, multiple pipelined classifiers, as well as the parallel processing, are discussed in this work, in order to provide a potential improvement for the future work. This thesis not only provides a successful synthesis of a face detection system on a hardware accelerator but also ignites intriguing ideas in terms of improvement aspects, such as approximating the design for finding an optimal energy-quality tradeoff corresponding to different applications as our future work. Our vision is thus to create a smart building system as a case study, capable of sensing our surrounding environment, and more important, directing different types of sensor data to the optimal place, in terms of computing devices, for analysis and making decisions autonomous at the proximity of the network edge to improve data privacy, latency, and bandwidth usage.
... The coordinate 'L' represents the lightness of color, while 'a' and 'b' denotes the chromatic range of green-magenta and blue-yellow, respectively. Because three coordinates are measured independently, it permits the measurement of infinitely many possible colors in three-dimensional real number space [25]. ...
Background and aim: All-ceramic prosthesis is widely used in modern dental practice because of its improved physico-mechanical and optical properties. These restorations are exposed to coloring agents from various nutrition and beverages in the oral cavity. Long-term color stability is critical for the success of these restorative materials. The purpose of this in vitro study was to assess the effect of common beverages and mouthwash on the color stability of lithium disilicate (LD), monolithic zirconia (MZ) and bilayer zirconia (BZ) surfaces. Material and methods: Thirty disc-shaped specimens from each material were fabricated; each group was subdivided (n = 10) according to coffee, green tea and chlorhexidine immersion solutions. The baseline color of ceramic discs was recorded according to the CIE L*a*b* system with a portable spectrophotometer. The second measurement was recorded after 3000 thermocycling and immersion in coloring agents for 7 days. The mean color difference was calculated and data were compared with Kruskal-Wallis and Mann-Whitney post hoc tests (0.05). Results: ΔE values for LD with the immersion of coffee, tea, and Chlorhexidine gluconate (CHG) were 1.78, 2.241 and 1.58, respectively. Corresponding ΔE values for MZ were 5.60, 5.19, and 4.86; marginally higher than the clinically acceptable level of 3.5. Meanwhile, BZ showed better color stability compared to MZ with ΔE values of 4.22, 2.11 and 1.43. Conclusion: Among the ceramics evaluated, LD ceramic was found to be more color stable, while MZ ceramics displayed a higher susceptibility to discoloration. MZ and BZ ceramic colors were significantly altered with coffee immersion, while LD ceramics were more affected by green tea.
... One of the major issues in using skin color in skin detection is how to choose a suitable color space. Numerous color models (RGB, CMY, and CMYK [5]; Hue, Saturation, and Intensity (HIS) [6,7]; Hue, Saturation, and Value (HSV) [8]; Normalized RGB [9]; and YCbCr [6,8] ) are used today because color science is a broad field encompassing many areas of applications. Many skin models have been developed based on color feature only using RGB color model [10], but these approaches are not robust enough to handle different lighting conditions and complex backgrounds containing surfaces and objects with skin-like colors. ...
Abstract Various approaches of skin detection have yet to demonstrate a stable state of performance. This is due to skin color in an image that is sensitive to variant illumination, camera adjustments, and human skin types. To contribute in overcome this problem a robust skin detection method that integrates both color and texture features is proposed. Texture features were estimated using statistical measures as range, standard deviation, and entropy. Back-propagation artificial neural network is then used to learn features and classify any given inputs. In this work, two skin detectors based on texture features only, and a combination of both color and texture features (proposed) have been constructed. Furthermore, the paper analyzes and compares the obtained results from the both skin detectors to show the impact of the integrating color and texture features to the robustness level. It found that the proposed skin detection method achieved a true positive rate of approximately 94.5% and a false positive rate of approximately 0.89%. Experimental results showed that proposed approach is more efficient compared with other state-of-the-art texture-based skin detector approaches.
... The aim of color segmentation is to separate out skin areas from non-skin regions in the RGB image. Use Gray World Assumption to get the color balanced RGB image to convert RGB image into YCbCr image [11]. ...
... These algorithms usually use YCbCr [8][9][10][11], RGB [7], HSV [12], IHLS [6] and CIELAB [6] spaces. Sometimes authors compare thresholding in different spaces such as [13] which compares thresholding in RGB,HSI and YCbCr. Threshold-based methods are mostly easy to train, take low processing time and are easy to execute. ...
skin detection is one of the most important targets of image processing and computer vision. One big concern about skin detection algorithms is their simplicity while keeping a good accuracy in discriminating skin and non-skin pixels. This paper presents a novel and robust skin detector. In this study, statistical information of each pixel and its neighbours were taken into account in order to deal with this concern. In proposed method, a cascaded classifier using AdaBoost algorithm was trained. Its errors were found and corrected. Finally, two edge detectors were used to make the algorithm more accurate. Also, some small tricks were used to make the whole process faster. The performances of proposed skin detector were evaluated using SFA skin database. Eventually, the method was compared with some popular and newly established skin detectors. The experimental results show that the proposed scheme outperforms other skin detection methods due to high precision and good recall.
... RGB color space system[2] ...
... RGB model is not ideal since the red, green and blue colour components are highly correlated. Skin colour region is more effectively extracted in YCbCr colour space because Cb and Cr have some distinct colour range for skin region [26]. Thus algorithm works quite well. ...
Real time video processing found its range of applications from defence to consumer electronics for surveillance, video conferencing etc. With the advent of FPGAs, flexible Real-Time Video Processing System (RTVPS) which can meet hard real-time constraints are easily realised with short development time. A hardware software co-design for an FPGA based real time video processing system to convert video in standard PAL 576i format to standard video of VGA / SVGA format with little utilisation of resources is realised and evaluated. Switching between multiple video streams, character/ text overlaying, skin colour detection is also incorporated. The system is also adaptable for rugged applications. VHDL codes for the architecture were synthesized using ALTERA Quartus II and targeted for ALTERA STRATIX I FPGA. The evaluated results show that the resource utilization is low for this design. Since system is also flexible, latest applications can be incorporated in future.
... Some authors have used other approaches to homogenize detected skin area in faces or human body, using median filter 7 or color quantization 18 in training images. Singh et al. 29 separated skin color in 1100 images using three color spaces and combining that with the conjunctive "and". They used a very similar idea to aggregation presented here, but in a boolean sense and using a thousand of images. ...
In this paper, a framework for detection of human skin in digital images is proposed.
This framework is composed of a training phase and a detection phase. A skin class
model is learned during the training phase by processing several training images in a
hybrid and incremental fuzzy learning scheme. This scheme combines unsupervised- and
supervised-learning: unsupervised, by fuzzy clustering, to obtain clusters of color groups
from training images; and supervised to select groups that represent skin color. At the
end of the training phase, aggregation operators are used to provide combinations of
selected groups into a skin model. In the detection phase, the learned skin model is used
to detect human skin in an efficient way. Experimental results show robust and accurate
human skin detection performed by the proposed framework.
Keywords: Fuzzy learning; color classification; skin detection; aggregation operators.
... To be able to generate ultimate colour, the spectral complements of these colours are mixed, and a 3-dimensional cube results, with three perpendicular axes representing R, G, and B or the RGB model. R, G, or B colours can have a value between 0 and 255 in a colour palette in each pixel [47]. R, G, B is identified as skin if [48]; "R > ...
Biometry based authentication and recognition have attracted greater attention due to numerous applications for security-conscious societies, since biometrics brings accurate and consistent identification. Face biometry possesses the merits of low intrusiveness and high precision. Despite the presence of several biometric methods, like iris scan, fingerprints, and hand geometry, the most effective and broadly utilized method is face recognition, because it is reasonable, natural, and non-intrusive. Face recognition is a part of the pattern recognition that is applied for identifying or authenticating a person that is extracted from a digital image or a video automatically. Moreover, current innovations in big data analysis, cloud computing, social networks, and machine learning have allowed for a straightforward understanding of how different challenging issues in face recognition might be solved. Effective face recognition in the enormous data concept is a crucial and challenging task. This study develops an intelligent face recognition framework that recognizes faces through efficient ensemble learning techniques, which are Random Subspace and Voting, in order to improve the performance of biometric systems. Furthermore, several methods including skin color detection, histogram feature extraction, and ensemble learner-based face recognition are presented. The proposed framework, which has a symmetric structure, is found to have high potential for biometrics. Hence, the proposed framework utilizing histogram feature extraction with Random Subspace and Voting ensemble learners have presented their superiority over two different databases as compared with state-of-art face recognition. This proposed method has reached an accuracy of 99.25% with random forest, combined with both ensemble learners on the FERET face database.
... Due to the high correlation between color components: red, green and blue, as each component is subject to the effect of luminance of the light intensity of the environment, so that suffers dissatisfaction on the part of many image processing applications. In practical, this model is not well suited to describe colors in terms of human interpretation [3]. ...
Human face states the inner emotions, thoughts and physical disorders. These emotions are expressed on the face via facial muscles. Researches indicate that facial expressions are the best way to expresse emotions. Human face expressions and micro expressions could be studied in the images and digital video frames. The estimated time through which a facial expression occurs is between 0.5 to 4 seconds, and a micro expression between 0.1-0.5. Also, in some references this value is stated 1.3, 1.15 and 1.25 seconds. Obviously, for the purpose of recording micro expressions, obtaining videos frames between 30 up to 200 fps is essential. Before depth sensors emerge, Facial Expressions Recognition (FER) was done only by Color images; But after depth sensors emerged and due to more data (depth dimension), recognition rate in FER increased significantly. This was tangible for a decade in this field. Facial expressions recognition has application in: human - computer (robot) interaction, 2D, 3D animation, psychology, non-verbal communication or body language, emotion recognition, security issues such lie detection, etc. Features which are used in this study are Histogram of Oriented Gradient (HOG), Gabor Filter, Speeded Up Robust Features (SURF), Local Phase Quantization (LPQ), Local Binary Pattern (LBP) and Haar Feature. Due to the shortage of RGB-D FER database, and also due to available databases weaknesses, a database including 40 individuals or subjects in a variety of age and genders with the Kinect V.2 sensor is gathered which to the extent acceptable has resolved the available databases with similar features weaknesses. On the other hand, it can be said this is the first depth database for Facial Micro Expression Recognition (FMER). This database is named Iranian Kinect Face Database (IKFDB). Considering that Kinect’s acquired data is divided into color and Depth parts, a hybrid feature extraction method for depth data, and based on pixel distance alterations with depth sensor is considered. Sections under the titles of age estimation and gender recognition is considered too. Also, a face detection and extraction algorithm out of Depth images is presented. These methods are evaluated and compared with the benchmark databases, and the proposed database. Databases used for evaluation are Eurecom Kinect Face DB, VAP RGBD Face DB, VAP RGBD-T Face, JAFFE, IKFDB, Face Grabber DB, Curtin Face, FEEDB, and CASME which are prepared in two kinds (image and video frames) by different RGB (color), Depth and Thermal sensors. Finally selected features, in the shape of feature vector, and for learning process, are sent to Support Vector Machine (SVM) and Multi-Layer Neural Network (MLNN) classifiers. The results are really satisfactory, and it indicates classification accuracy improvement in some databases and methods. Also some of these actions are performed on some of these databases for the first time.
In this paper we are going to develop a robust automated web application for transacting money in higher level security purpose with high facial recognition. First we have to register our personal details with our face and pass the liveness detection test. Haar cascade-based algorithm has been applied for fast and simple face detection from the input image. The face image is then being converted into grayscale image. After that, the iris, eyebrows, nose, mouth of candidates are extracted from the intensity valleys from the detected face it will consider as datasets now our system can understand who are all authorized and unauthorized. While login our face will be recognized if we authorized will can move to authorized page else our system won’t allow to login. For higher security purpose we have used face recognition module. During the transaction our face will be recognized it will allow only the authorized account holder to transact a money, it doesn’t allow others to transact a money.
With the phenomenal growth of video and image databases, there is a tremendous need for intelligent systems to automatically understand and examine information, as doing so manually is becoming increasingly difficult. The face is important in social interactions because it conveys information. Detecting a person's identity and feelings Humans do not have a great deal of ability to identify. Machines have different faces. As a result, an automatic face detection system is essential.in face recognition, facial expression recognition, head-pose estimation, and human–computer interaction, and so on Face detection is a computer technology that determines the location and size of a person's face. It also creates a digital image of a human face. Face detection has been a standout topic in the science field This paper provides an in-depth examination of the various techniques investigated for face detection in digital images. Various face challenges and applications. This paper also discusses detection. Detection features are also provided. In addition, we hold special discussions on the practical aspects of developing a robust face detection system, and finally. This paper concludes with several promising research directions for the future.
In today’s world, technologies like virtual and augmented reality grow every second and provide diverse dimensions to various other domains. These technologies are more likely to bring users into the virtual world to help them understand multiple concepts more closely. In fact, in the present era, little kids of the current generation are more curious about the know-how of things in a more visual way. The curiosity among the young kids to understand various concepts requires a better learning environment. The kids may explore the working of the objects in 3D by visualizing the natural movements. But practically, this demands a virtualized environment for them to understand these things and that too in a safe environment. Now, this is somehow getting possible due to the latest advancement within these technologies which opens space for exploration for various domains. Recently Google has developed ARCore, an SDK that allows the development of augmented reality-based applications. In this virtualized environment, the user can immerse themselves even when they are not physically present. In our model, we have facilitated the user to explore different dimensions of the computer vision which are required for understating various concepts in a very cost-effective manner. In our model, we have provided a vision-based interface to the user that replaces peripheral hardware without affecting their performances and facilitates them to use these interfaces embedded with AR to better understand 3D concepts.KeywordsOpenCVOpenGLComputer visionVirtual realityAugmented realityARCore
Recently, virtual reality (VR) has received significant attention among
researchers to accomplish realistic interaction between humans and computers.
Among the different application areas of VR, multimedia human computer interaction requires accurately collect and identify the motions of human body in real
time. For enabling highly realistic and efficient communication among humans and
computers, design of human motion recognition system becomes necessary to determine diverse complex and distinct human actions. Therefore, this study designs a
new optimal bidirectional long short term memory (BiLSTM) with fully convolution network (FCN), called OBiLSTM-FCN model for human motion recognition in VR environment. The proposed OBiLSTM-FCN model comprises different
processes namely feature extraction, classification, and hyperparameter optimization.
Primarily, kernel based linear discriminant analysis (LDA) approach is employed as
a feature extraction technique. In addition, the BiLSTM-FCN technique is applied
as a recognition model to determine human motions. Finally, the Adam optimizer is
applied to optimally tune the hyperparameters involved in the BiLSTM-FCN model.
The performance validation of the OBiLSTM-FCN model take place and the resultant
values portrayed the improved performance over the other compared methods
Uniqueness or individuality of an individual is his face. In this project face of an individual is used for the purpose of attendance making automatically. Attendance of the student is very important for every college, universities and school. Conventional methodology for taking attendance is by calling the name or roll number of the student and the attendance is recorded. Time consumption for this purpose is an important point of concern. Assume that the duration for one subject is around 60 minutes or 1 hour & to record attendance takes 5 to 10 minutes. For every tutor this is consumption of time. To stay away from these losses, an automatic process is used in this project which is based on image processing. In this project face detection and face recognition is used. Face detection is used to locate the position of face region and face recognition is used for marking the understudy's attendance. The database of all the students in the class is stored and when the face of the individual student matches with one of the faces stored in the database then the attendance is recorded.
In 1995, developers and programmers have begun focusing on developing websites and data protection. This came as a result of the increase and growth of many digital services such as e-shopping services, bank account management, e-mails, and many other services that require a high level of protection and security. As the penetration internet gowned it throws many challenges towards information protection. Hence the importance of providing protection and security for websites and users to protect their data and privacy, and bearing in mind that authentication is the first and basic step to protect and encrypt information sources, it also provides a safe environment for Internet users, which gives them comfort and security when dealing with these sites. The idea of our research aims for developing a security system that combines some recognition algorithms based on machine learning and deep learning techniques to verify the identity of each user (authentication) for the website and all the services associated with it. Thus, we can achieve the highest level of protection and security from the digital gaps that hackers may exploit. Therefore, the facial expression became the most important technique to identify hackers. The obtained results reveal that deep learning-based techniques for face recognition over a collected dataset are superior to conventional machine learning techniques.
Face recognition from the images is challenging due to the wide variability of face appearances and the complexity of the image background. This paper proposes a novel approach for recognizing the human faces. The recognition is done by comparing the characteristics of the new face to that of known individuals. It has Face localization part, where mouth end point and eyeballs will be obtained. In feature Extraction, Distance between eyeballs and mouth end point will be calculated. The recognition is performed by Neural Network (NN) using Back Propagation Networks (BPN) and Radial Basis Function (RBF) networks. The recognition performance of the proposed method is tabulated based on the experiments performed on a number of images.
Deteksi wajah merupakan tahapan awal yang penting dalam proses pengenalan wajah. Teknik-teknik pengenalan wajah yang dilakukan selama ini banyak yang menggunakan asumsi bahwa data wajah yang tersedia memiliki ukuran yang sama dan latar belakang yang seragam. Di dunia nyata, asumsi ini tidak selalu berlaku karena wajah dapat muncul dengan berbagai ukuran dan posisi di dalam citra dan dengan latar belakang yang bervariasi. Penelitian ini menggunakan kombinasi dari metode color-based dan feature-based. Terdapat dua tahapan utama dalam penelitian ini. Pertama, segmentasi kulit pada ruang warna YCbCr. Kedua, ekstraksi fitur wajah menggunakan rasio tinggi terhadap lebarkan di data wajah, pencarian area mulut dan pencarian area mata. Hasil dari penelitian ini, dari pendeteksian pada 45 citra warna dengan background kompleks didapat presentase keberhasilan sebesar 66,67%.
The main purpose of this research is to enhance the communication of the disabled community. The authors of this chapter propose an enhanced interpersonal-human interaction for people with special needs, especially those with physical and communication disabilities. The proposed model comprises of automated real time behaviour monitoring, designed and implemented with the ubiquitous and affordable concept in mind to suit the underprivileged. In this chapter, the authors present the prototype which encapsulates an automated facial expression recognition system for monitoring the disabled, equipped with a feature to send Short Messaging System (SMS) for notification purposes. The authors adapted the Viola-Jones face detection algorithm at the face detection stage and implemented template matching technique for the expression classification and recognition stage. They tested their model with a few users and achieved satisfactory results. The enhanced real time behaviour monitoring system is an assistive tool to improve the quality of life for the disabled by assisting them anytime and anywhere when needed. They can do their own tasks more independently without constantly being monitored physically or accompanied by their care takers, teachers, or even parents. The rest of this chapter is organized as follows. The background of the facial expression recognition system is reviewed in Section 2. Section 3 is the description and explanations of the conceptual model of facial expression recognition. Evaluation of the proposed system is in Section 4. Results and findings on the testing are laid out in Section 5, and the final section concludes the chapter.
Human skin detection is an important and challenging problem in computer vision. Skin detection can be used as the first phase in face detection when using color images. The differences in illumination and ranges of skin colors have made skin detection a challenging task. Gaussian model, rule based methods, and artificial neural networks are methods that have been used for human skin color detection. Deep learning methods are new techniques in learning that have shown improved classification power compared to neural networks. In this paper the authors use deep learning methods in order to enhance the capabilities of skin detection algorithms. Several experiments have been performed using auto encoders and different color spaces. The proposed technique is evaluated compare with other available methods in this domain using two color image databases. The results show that skin detection utilizing deep learning has better results compared to other methods such as rule-based, Gaussian model and feed forward neural network.
Extraction of human emotions from facial expression has attracted significant attention in computer vision community. There are several appearance based techniques like local binary patterns (LBP), local directional patterns (LDP), local ternary patterns (LTP) and gradient local ternary patterns (GLTP). Recently, many investigations have been done to improve these feature extraction techniques. Although GLTP has achieved an improvement in robustness to noise and illumination, it encodes image gradient in four directions and two orientations only. This paper proposes to improve GLTP to directional gradient local ternary patterns (DGLTP) by encoding image gradient on eight directions and four orientations. The eight directional Kirsch mask is used to encode the image gradient followed by dimensionality reduction using linear discriminant analysis (LDA) and AVG, MAX and MIN pooling techniques are compared for fusing facial expression features. The proposed technique was experimented on JAFFE facial expression dataset with support vector machine (SVM). The experimental results show that proposed technique improved accuracy of GLTP.
The purpose of this chapter is to discuss and present a computational framework for detecting and analysing facial expressions efficiently. The approach here is to identify the face and estimate regions of facial features of interest using the optical flow algorithm. Once the regions and their dynamics are computed a rule based system can be utilised for classification. Using this framework, we show how it is possible to accurately identify and classify facial expressions to match with FACS coding and to infer the underlying basic emotions in real time.
In this paper, we propose a new technique to recognize vessels in robot-assisted laparoscopic surgery images by using surgical instruments. The proposed method does not require additional hardware or parameter adjustment because it detects blood vessels by using only the color information of the image. The concept of a hessian matrix is used in the HSV color space of the image to detect the edges of the blood vessels. In addition, the histogram equalization technique, clustering technique, and region growing are used to remove the surgical tools. Images of actual robot-assisted laparoscopic surgery videos were used. The processing speed was approximately 0.3 s per frame at 640p and approximately 0.8 s per frame at 1280p. The average recall was 92.67%.
Computer vision provides a powerful tool for the interaction between man and machine. The barrier between physical objects (paper, pencils, calculators) and their electronic counterparts limits both the integration of computing into human tasks, and the population willing to adapt to the required input devices. Computer vision, coupled with video projection using low cost devices, makes it possible for a human to use any convenient object, including fingers, as digital input devices. In such an “augmented reality”, information is projected onto ordinary objects and acquired by watching the way objects are manipulated. In the first part of this paper we describe experiments with techniques for watching the hands and recognizing gestures.Vision of the face is an important aspect of human-to-human communication. We have been experimenting with the use of computer vision “watch the face”. In the second part of this paper we describe techniques for detecting, tracking and recognizing faces. When combined with real time image processing and active control of camera parameters, these techniques can greatly reduce the communications bandwidth required for videophone and videoconference communications.