Conference Paper

Skin Segmentation Using Color Distance Map and Water-Flow Property

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

A new approach for skin region segmentation is proposed. It uses color distance map (CDM) and an algorithm based on the property of flow of water. The CDM itself is a grayscale image, which makes the algorithm very simple. However, it is still capable of providing color information based on which some skin and non-skin seed regions can be determined reliably. Then a water-flow based procedure determines skin and non-skin segments completely. The color distance map is robust against variations in imaging conditions and the water-flow procedure efficiently uses the region information to extract solid skin segments without generating much noisy segments.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Paper in [85] came up with a model which can overcome the sensitivity in regards with the variations in the condition of lighting and also while complex backgrounds used Hue, Saturation, and Value color space and multi-level segmentation. Furthermore in [86] The author proposed a new method in skin region segmentation, his method utilize color distance map (CDM) and another algorithm which is based on water flow. Paper number fourteen [87] come up with framework was introduced, the content-based framework was aimed for detecting skins with grayscale images. ...
... Again, it may be precise but lacking in simplicity, or even it may even be complicated and slow [84]. In majority of applications, a particular level of accuracy and speed is adequate, and the complexity of a model is determined by this [86]. ...
Article
Full-text available
This study review and analysis the literature on skin detector (SD), in order to establish the coherent taxonomy and figure out the gap on this pivotal research area. An extensive search is conducted to identify articles that deal with skin detection, skin segmentation, skin tone detector and skin recognition issues, related techniques are reviewed comprehensively and a coherent taxonomy of these articles is established. ScienceDirect, IEEE Xplore and Web of Science databases are checked for articles on skin detector. A total of 2803 papers are collected from 2007 to February 2018. The set comprised 173 articles. The largest portion of the papers (n=158/173) = 91% belong to Development and Design, that is aimed to develop an approach for skin classifier into skin and non-skin. A sum total of (n=5/173)=3% of the papers belong to Evaluation and Framework, (n=10/173) = 6% papers was categorized as Comparative Study. This study discusses the open challenges, motivations and recommendations of the related works. Furthermore, state-of-the-art is a step to demonstrate the novelty of the presented study by conducted a statistical analysis for previous studies such as (Dataset, Colour spaces, features, image type, and Classification techniques) as a future direction for other researchers who are interested in skin detector (SD).
... The major category which has a high threat B R. Balamurali balamurale@gmail.com A. Chandrasekar drchandrucse@gmail.com 1 Chennai Institute of Technology, Anna university, Chennai, India 2 CSE, St Joseph's College of Engineering, Chennai, India to pornography is the educational community [6]. The introduction of android devices and applications related to internet services has exaggerated the situation [4]. ...
... ZhiweiJiang [10] tried to combine the RGB color space with texture values in Grayscale to neglect the effect of illumination. In the work of Abdullah-Al-Wadud [1] segmentation is done on the image and color distance map is used to determine skin pixel region and non-skin pixel regions reliably. However, Phung [16], in his work reveals that the results based on segmentation fails under chrominance channels. ...
Article
Full-text available
Multi-parameter algorithm with statistical approach for adult image discretion is proposed to figure out the obscene images. In this work, we propose an analysis on different color spaces to identify an effective pixel identification for human skin. In this work, an algorithm is incorporated to verify and spirit away the unambiguous image by identifying high skin pixel rate. This framed algorithm is verified in terms of accuracy, true negatives and false positives and the results expressed in this paper show that the algorithm worked well and fast in detecting obscene images.
... However, segmentation performance degrades when only chrominance channels are used in classification. In chrominance based methods, some valuable skin colour information will be lost whilst attempting to separate luminance from chrominance according to [10]. Shin et al. [11] question the benefit of colour transformation for skin tone detection, e.g., RGB and non-RGB colour spaces. ...
... The b component has the least representation of skin colour and therefore it is normally omitted in skin segmentation [29]. Abdullah-Al-Wadud and Chae [10] use a colour distance map (CDM) applied to RGB colours, although that can be extended to any colour space. They implement an algorithm based on the property of the flow of water to further refine the output using an edge operator. ...
Article
Challenges face biometrics researchers and particularly those who are dealing with skin tone detection include choosing a colour space, generating the skin model and processing the obtained regions to fit applications. The majority of existing methods have in common the de-correlation of luminance from the considered colour channels. Luminance is underestimated since it is seen as the least contributing colour component to skin colour detection. This work questions this claim by showing that luminance can be useful in the segregation of skin and non-skin clusters. To this end, here we use a new colour space which contains error signals derived from differentiating the grayscale map and the non-red encoded grayscale version. The advantages of the approach are the reduction of space dimensionality from 3D, RGB, to 1D space advocating its unfussiness and the construction of a rapid classifier necessary for real time applications. The proposed method generates a 1D space map without prior knowledge of the host image. A comprehensive experimental test was conducted and initial results are presented. This paper also discusses an application of the method to image steganography where it is used to orient the embedding process since skin information is deemed to be psycho-visually redundant.
... This process is repeated for new adjacent pixels until no neighbouring pixels satisfy the conditions and the growth of the region stops. Variations of this method have been proposed in [34][35][36] where different measurement techniques have been proposed to measure the similarity between adjacent pixels for assigning them to a region, such as Euclidean distance, colour distance map or probability measuring. ...
Article
Full-text available
Human skin segmentation is a crucial task in computer vision and biometric systems, yet it poses several challenges such as variability in skin colour, pose, and illumination. This paper presents a robust data-driven skin segmentation method for a single image that addresses these challenges through the integration of contextual information and efficient network design. In addition to robustness and accuracy, the integration into real-time systems requires a careful balance between computational power, speed, and performance. The proposed method incorporates two attention modules, Body Attention and Skin Attention, that utilize contextual information to improve segmentation results. These modules draw attention to the desired areas, focusing on the body boundaries and skin pixels, respectively. Additionally, an efficient network architecture is employed in the encoder part to minimize computational power while retaining high performance. To handle the issue of noisy labels in skin datasets, the proposed method uses a weakly supervised training strategy, relying on the Skin Attention module. The results of this study demonstrate that the proposed method is comparable to, or outperforms, state-of-the-art methods on benchmark datasets.
... This process is repeated for new adjacent pixels until no neighboring pixels satisfy the conditions and the growth of the region stops. Variations of this method have been proposed in [34,35,36] where different measurement techniques have been proposed to measure the similarity between adjacent pixels for assigning them to a region, such as Euclidean distance, color distance map or probability measuring. ...
Preprint
Full-text available
Human skin segmentation is a crucial task in computer vision and biometric systems, yet it poses several challenges such as variability in skin color, pose, and illumination. This paper presents a robust data-driven skin segmentation method for a single image that addresses these challenges through the integration of contextual information and efficient network design. In addition to robustness and accuracy, the integration into real-time systems requires a careful balance between computational power, speed, and performance. The proposed method incorporates two attention modules, Body Attention and Skin Attention, that utilize contextual information to improve segmentation results. These modules draw attention to the desired areas, focusing on the body boundaries and skin pixels, respectively. Additionally, an efficient network architecture is employed in the encoder part to minimize computational power while retaining high performance. To handle the issue of noisy labels in skin datasets, the proposed method uses a weakly supervised training strategy, relying on the Skin Attention module. The results of this study demonstrate that the proposed method is comparable to, or outperforms, state-of-the-art methods on benchmark datasets.
... A skin tone detection algorithm for an adaptive approach to steganography by Abbas Cheddad , Joan Condell, Kevin Curran, Paul Mc Kevitt [5] at 2009 Detecting human skin tone is of utmost significance in severa applications inclusive of, video surveillance, face and gesture recognition, human computer interplay, human pose modelling, image and video indexing and retrieval, image enhancing, automobile drivers' drowsiness detection, controlling users' surfing behaviour (e.G., browsing indecent sites) and steganography. Skin Segmentation Using Color Distance Map and Water-flow Property by M. Abdullah-Al-Wadud, Oksam Chae [6] at 2008. A new technique for pores and skin place segmentation is proposed. ...
... Abdullah-Al-Wadude et al. [46] also proposed an algorithm uses color distance map (CDM); a gray scale image robust against variations in imaging conditions , and an algorithm based on the property of flow of water which uses spatial analysis to extract skin blobs . Recently, in order to overcome leakage issue, Kawulok et al. [47,48] proposed an energy based approach in which the probability of pixels are utilized to determine the skinness. ...
Article
Skin Segmentation is widely used in biometric applications such as face detection, face recognition, face tracking, and hand gesture recognition. However, several challenges such as nonlinear illumination, equipment effects, personal interferences, ethnicity variations, etc., are involved in detection process that result in the inefficiency of color based methods. Even though many ideas have already been proposed, the problem has not been satisfactorily solved yet. This paper introduces a technique that addresses some limitations of the previous works. The proposed algorithm consists of three main steps including initial seed generation of skin map, Otsu segmentation in color images, and finally a two-stage diffusion. The initial seed of skin pixels is provided based on the idea of ternary image as there are certain pixels in images which are associated to human complexion with very high probability. The Otsu segmentation is performed on several color channels in order to identify homogeneous regions. The result accompanying with the edge map of the image is utilized in two consecutive diffusion steps in order to annex initially unidentified skin pixels to the seed. Both quantitative and qualitative results demonstrate the effectiveness of the proposed system in compare with the state-of-the-art works.
... The blockdiagram of the segmentation process is shown inFig. 2. The first sclera map is created by classifying each pixel into skin or not-skin labels using the Color Distance Map (CDM)[9]. First, two skin clusters for natural illumination and flash illumination conditions are extracted which are defined as ...
Conference Paper
Full-text available
Sclera recognition has provoked research interest recently due to the distinctive properties of its blood vessels. However, segmenting noisy sclera areas in eye images under relaxed imaging constraints, such as different gaze directions, capturing on-the-move and at-a-distance, has not been extensively investigated. In our previous work, we proposed a novel method for sclera segmentation under unconstrained image conditions with a drawback being that the eye gaze direction is manually labeled for each image. Therefore, we propose a robust method for automatic eye corner and gaze detection. The proposed method involves two levels of eye corners verification to minimize eye corner point misclassification when noisy eye images are introduced. Moreover, gaze direction estimation is achieved through the pixel properties of the sclera area. Experimental results in on-the-move and at-a-distance contexts with multiple eye gaze directions using the UBIRIS.v2 database show a significant improvement in terms of accuracy and gaze detection rates.
... However, this may be a time-consuming procedure as it involves simulated annealing for every analyzed image. M. Abdullah-Al-Wadud [2] proposed to transform an image into a single-dimensional color distance map, in which a waterflow procedure is carried out to segment skin regions. A.Y. Dawod used basic edge detectors to determine the boundaries of skin regions [22]. ...
... Abdullah-Al-Wadude et al. [192] also proposed an algorithm that uses color distance map (CDM); a gray scale image robust against variations in imaging conditions and an algorithm based on the property of flow of water which uses spatial analysis to extract skin blobs. For spatial analysis, the edge map of the image is utilized as a height of a pixel and then water is dropped towards both sides of the edges. ...
... Also, using GMM with any number of kernels and any threshold is not reliable for finding seed points. Abdullah-Al-Wadude et al. [1] also proposed an algorithm uses color distance map (CDM); a gray scale image robust against variations in imaging conditions, and an algorithm based on the property of flow of water which uses spatial analysis to extract skin blobs. Recently, in order to overcome leakage issue, Kawulok et al. [22] proposed an energy based approach in which the probability of pixels are utilized to determine the skinness. ...
Article
Full-text available
Recently, skin detection has been employed in multifarious applications of computer vision including face detection, gesture recognition, etc. This is mainly due to the appealing characteristics of skin color and its potency to segment objects. However, there are certain challenges involved in utilizing human complexion as a feature to detect faces, and they have led to the inefficiency of many methods. In order to counteract these factors, in this paper, a skin segmentation method which exploits a multi step diffusion algorithm to detect skin regions is presented. The method starts with conservative extraction of skin seeds in each frame which is accomplished by using fusion of ternary-based human motion detection, modified Bayesian classifier, and a feedback mechanism. Subsequently, these candidate skin pixels are utilized in a 2-stage diffusion scheme to detect other skin pixels. Both quantitative and qualitative results demonstrate the effectiveness of the proposed system in comparison with other works.
... This method creates two binary maps for the eye image and fuses them to enhance the sclera area detection for noisy images. The block diagram of the segmentation process is shown in Fig. 2. The first sclera map is created by classifying each pixel into skin or not-skin labels using the Color Distance Map (CDM) [9]. First, two skin clusters for natural illumination and flash illumination conditions are extracted which are defined as ...
... However, this may be a time-consuming procedure as it involves simulated annealing for every analyzed image. M. Abdullah-Al-Wadud [2] proposed to transform an image into a single-dimensional color distance map, in which a waterflow procedure is carried out to segment skin regions. A.Y. Dawod used basic edge detectors to determine the boundaries of skin regions [22]. ...
Chapter
This chapter presents an overview of existing methods for human skin detection and segmentation. First of all, the skin color modeling schemes are outlined, and their limitations are discussed based on the presented experimental study. Then, we explain the techniques which were reported helpful in improving the efficacy of color-based classification, namely (1) textural features extraction, (2) model adaptation schemes, and (3) spatial analysis of the skin blobs. The chapter presents meaningful qualitative and quantitative results obtained during our study, which demonstrate the benefits of exploiting particular techniques for improving the skin detection outcome.
Research
Full-text available
Steganography is the science that involves communicating secret data in an appropriate multimedia carrier like image, video and audio files. it does not replace cryptography but rather boots the security using its obscurity features. When we work on a network the security requirements of a user as well as a network increases. There are number of available ways over the network to achieve the information security. Till now the available methods hide the secret data over the image on a fixed pattern that makes a user identify the pattern easily. We are providing a dynamic pattern extraction approach using biometric. It advocates an object-oriented approach in which skin tone detected areas in the image are selected for embedding where possible. After detecting the skin area, edge detection is performed by using the canny edge detection method. As the edges will be detected we will use this area as the pattern to hide the data over the image. The secret data is compressed using the DWT technique and then further compressed secret information is encrypted using RSA algorithm with bit shift method. This proposed technique provides more security to the data as data is embedded not in the whole image but only at some specified location. Due to this image distortion of the carrier image is less and it is difficult to identify the Secret image.
Article
Full-text available
The data explosion caused by the Internet and its applications has given researchers immense scope for data analysis. A large amount of data is available in form of images. Image processing is required for better understandability of an image. Various image processing steps are available for improving the image in different application areas. Various applications like medical imaging, face recognition, biometric security, fruit quality evaluation, and traffic surveillance depend only on image and its analysis. This analysis in several applications is highly dependent on the outcome of image segmentation. This paper focuses on the good segmentation of different kinds of fruits through multi-level thresholding. In this paper, multi-level segmentation based on modified Firefly Algorithm (FA) with Kapur’s, Tsallis, and fuzzy entropy is proposed. FA is used to optimize fuzzy parameters for obtaining optimal thresholds. The levy flight and local search are implemented with FA. The various experiments have been performed on apple, banana, mango, and orange images with the distinct threshold (i.e., 2, 3, 4) values. The proposed algorithm has been estimated quantitatively and qualitatively by using parameters like peak signal-to-noise ratio (PSNR) and structured similarity index metric (SSIM).
Article
Full-text available
The data explosion caused by the Internet and its applications has given researchers immense scope for data analysis. A large amount of data is available in form of images. Image processing is required for better understandability of an image. Various image processing steps are available for improving the image in different application areas. Various applications like medical imaging, face recognition, biometric security, and traffic surveillance, etc. depend only on image and its analysis. This analysis in several applications is highly dependent on the outcome of image segmentation. This paper focuses on good segmentation through multi-level thresholding. In this research, the algorithm includes two modules related to Entropy and variance. The first module is concerned with the modified firefly algorithm (FA) with Kapur’s, Tsallis, and Fuzzy Entropy. FA is used to optimize fuzzy parameters for obtaining optimal thresholds. The second module is derived from the principle of variance between two classes known as between variance or inter-cluster variance. The opposition-based the learning method is used for initializing the population of candidate solutions and levying flight and local search is implemented with FA. The various experiments have been performed on Berkeley and benchmark images with distinct threshold (i.e. 2, 3, 4, 5) values. The proposed algorithm has been estimated and compared with known metaheuristic optimization methods like particle swarm optimization (PSO) and electromagnetism optimization (EMO). The results have been assessed quantitatively and qualitatively by using parameters like Peak signal-to-noise ratio (PSNR), structured similarity index metric (SSIM), objective function values, and convergence curve. The algorithm proposed observed better experiment results than PSO, EMO in terms of persistency and quality.
Conference Paper
Sclera is the white outer layer of the eye surrounding the cornea. Nowadays, the applications of image processing techniques on sclera are widely used in various fields, such as pattern recognition technique, biometrics, machine learning, image analysis and a range of medical applications, beneficial to humanity. Segmentation is one of the various technique in image processing. In this paper, we propose a simple technique by using thresholding method to segment the sclera images. The segmentation algorithm was implemented on RGB images. The aim of this paper is to segment the sclera by using thresholding method. Global thresholding and color thresholding were introduced in this system for segmenting sclera image. The result showed that the proposed segmentation technique has successfully segmented the sclera images by removing the nonessential parts and retaining the particular region of interest. The proposed method showed that the segmentation algorithm was able to segment the sclera images with a promising percentage of similarity indexes.
Article
The paper aims at classifying the extend of severity of the diseases that affect the human eye and finding out the symptoms of certain other diseases that affect the human body. Nowdays people worry more about losing their sight than about losing their memory and their ability to walk or hear. Many people are unaware of the warning signs of eye diseases and other diseases that can be suspected by examining the eye. In this paper we look at some of the most common diseases that can be suspected by closely looking and analyzing data that can be obtained from the sclera region of the eye. The image can be acquired using an ordinary camera of at least 8 mega pixels resolution. This technique of diagnosis saves the time of a medical practitioner and also enables diagnosis from distant places.
Article
The vessel structure as viewed on the sclera, the white and opaque outer protective covering of the eye, is stable over time and unique to each person. A few researchers have studied colour changes that occur in the sclera region and vein pattern deformities and have reported promising, but low accuracy, initial results. The vessel pattern on the top of the sclera poses several challenges: the vein structure moves and deforms with the movement of the eye and its surrounding tissues; images of sclera patterns are often defocused and/or saturated; and, most importantly, the vein structure in the sclera is multi-layered and has complex non-linear deformation. The colour of the conjunctiva layer above the sclera gives a very good indication of the well being of a person. The pattern of the sclera based blood vessels is a good database for the identification of several diseases and the thickness of the vessels give an indication of the mental state of the person. This makes the scanning and storing of the sclera vein pattern and the colour of the sclera a very good database for monitoring the physical and mental state of a person.
Article
With the development of technology concerning face recognition, it has aroused great attentions from related scientists and is envisioned to be widely applied in various fields in the future. In this paper we are going to describe an effective way for detecting face in the statistic images. The proposed method will firstly detect face region via using the color model, which has been described in my previous paper [1] so as to locate the candidate pixels. Then we analyze the gray value distribution of candidate pixels to segment a more accurate region. Finally the face region is likely to be obtained after the exclusion of superfluous skin region.
Article
The blood vessel structure of the sclera is unique to each person, and it can be remotely obtained nonintrusively in the visible wavelengths. Therefore, it is well suited for human identification (ID). In this paper, we propose a new concept for human ID: sclera recognition. This is a challenging research problem because images of sclera vessel patterns are often defocused and/or saturated and, most importantly, the vessel structure in the sclera is multilayered and has complex nonlinear deformations. This paper has several contributions. First, we proposed the new approach for human ID: sclera recognition. Second, we developed a new method for sclera segmentation which works for both color and grayscale images. Third, we designed a Gabor wavelet-based sclera pattern enhancement method to emphasize and binarize the sclera vessel patterns. Finally, we proposed a line-descriptor-based feature extraction, registration, and matching method that is illumination, scale, orientation, and deformation invariant and can mitigate the multilayered deformation effects and tolerate segmentation error. The experimental results show that sclera recognition is a promising new biometrics for positive human ID.
Article
The paper describes ongoing research in computer vision based gestural interfaces. The aim of the research is to develop a novel, multiuser gesture recognition system that addresses key factors of human computer interaction; the latter is not fully addressed in computer vision research. The focus of current work has been on efficient and accurate skin colour detection. Skin colour based segmentation for computer vision applications involving human users has received considerable attention over the years. One of the popular techniques is physics based skin colour locus in chromaticity space. The skin distribution in chromaticity space has shown robustness to changing illumination conditions. The paper proposes a technique based on skin samples for identifying skin locus, this does not require knowledge of camera parameters and correlated colour temperature of the illuminant. The paper also evaluates the potential for using skin locus based segmentation for an indoor gesture recognition system; previous studies have mainly focussed on face detection and face tracking.
Article
Iris recognition is tested to be one of the most reliable approaches for automatic personal recognition under the infrared light. However, it is challenging and difficult to acquire good quality iris patterns of dark color eyes in the visible wavelengths. Sclera recognition can achieve good recognition accuracy in the visible spectrum, but the performance will drop when the sclera pattern is saturated. In this paper, we proposed a comprehensive multimodal eye recognition system that uses both iris and sclera patterns for recognition from the same eye image. The experimental results show that the proposed multimodal frontal eye recognition method can achieve better recognition accuracy than unimodal iris or sclera recognition. In addition, we propose a multimodal multi-angle eye recognition system that can further improve the recognition accuracy.
Article
Multimodal biometrics use more than one means of biometric identification to achieve higher recognition accuracy, since sometimes a unimodal biometric is not good enough used to do identification and classification. In this paper, we proposed a multimodal eye recognition system, which can obtain both iris and sclera patterns from one color eye image. Gabor filter and 1-D Log-Gabor filter algorithms have been applied as the iris recognition algorithms. In sclera recognition, we introduced automatic sclera segmentation, sclera pattern enhancement, sclera pattern template generation, and sclera pattern matching. We applied kernelbased matching score fusion to improve the performance of the eye recognition system. The experimental results show that the proposed eye recognition method can achieve better performance compared to unimodal biometric identification, and the accuracy of our proposed kernel-based matching score fusion method is higher than two classic linear matching score fusion methods: Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA).
Conference Paper
Full-text available
It is well known that the color of many natural and man-made objects is often very similar to that of human skin, such as sand, brick, to name a few. For the task of skin detection, it is often a very challenge task to identify the right skin locations while being robust against the distraction of these objects. In this paper, we present an on-line learning approach to model human skin by utilizing the similarity between neighboring pixels, and then combine it with region growing technique to accurately segment skin regions. By assuming the colors of neighboring skin pixels in YCbCr color space follows conditional Gaussian distributions, an on-line learning update is devised to efficiently estimate the parameters of these Gaussian distributions. For the inference stage, our algorithm first evaluates the color distance map in RGB space to reliably place the seeds of skin regions, then segment the skin regions by iterative seed growing based on the learned skin models. Empirical evaluations demonstrate the efficacy of the proposed approach.
Conference Paper
Now-a-days digital camera have become very inexpensive as well as available device to common people and they usually like to see them better in pictures, which requires some post processing. To fulfill this requirement we propose a framework that integrates face detection, facial feature segmentation, intelligent filter design and control mechanisms into a single processing framework that requires minimal user assistance. For the design, selection and control mechanisms of the filters, we adopt the blackboard based expert system architecture. For a specified task, the current conditions of different parts of the image are analyzed and then a specific filter or set of filters are chosen to be applied based on the knowledge accumulated from experts.
Article
Full-text available
The segmentation of skin regions in color images is a preliminary step in several applications. Many different methods for discriminating between skin and non-skin pixels are available in the literature. The simplest, and often applied, methods build what is called an "explicit skin cluster" classifier which expressly defines the boundaries of the skin cluster in certain color spaces. These binary methods are very popular as they are easy to implement and do not require a training phase. The main difficulty in achieving high skin recognition rates, and producing the smallest possible number of false positive pixels, is that of defining accurate cluster boundaries through simple, often heuristically chosen, decision rules. In this study we apply a genetic algorithm to determine the boundaries of the skin clusters in multiple color spaces. To quantify the performance of these skin detection methods, we use recall and precision scores. A good classifier should provide both high recall and high precision, but generally, as recall increases, precision decreases. Consequently, we adopt a weighted mean of precision and recall as the fitness function of the genetic algorithm. Keeping in mind that different applications may have sharply different requirements, the weighting coefficients can be chosen to favor either high recall or high precision, or to satisfy a reasonable tradeoff between the two, depending on application demands. To train the genetic algorithm (GA) and test the performance of the classifiers applying the GA suggested boundaries, we use the large and heterogeneous Compaq skin database.
Article
Full-text available
This paper describes a computational approach to edge detection. The success of the approach depends on the definition of a comprehensive set of goals for the computation of edge points. These goals must be precise enough to delimit the desired behavior of the detector while making minimal assumptions about the form of the solution. We define detection and localization criteria for a class of edges, and present mathematical forms for these criteria as functionals on the operator impulse response. A third criterion is then added to ensure that the detector has only one response to a single edge. We use the criteria in numerical optimization to derive detectors for several common image features, including step edges. On specializing the analysis to step edges, we find that there is a natural uncertainty principle between detection and localization performance, which are the two main goals. With this principle we derive a single operator shape which is optimal at any scale. The optimal detector has a simple approximate implementation in which edges are marked at maxima in gradient magnitude of a Gaussian-smoothed image. We extend this simple detector using operators of several widths to cope with different signal-to-noise ratios in the image. We present a general method, called feature synthesis, for the fine-to-coarse integration of information from operators at different scales. Finally we show that step edge detector performance improves considerably as the operator point spread function is extended along the edge.
Conference Paper
Full-text available
Techniques for colour-based tracking of faces or hands often assume a static skin colour model. However, skin color perceived by a camera can change when lighting changes. In common real environments multiple light sources impinge on the skin. Therefore, for robust skin pixel detection, a dynamic skin colour model that can cope with the changes must be employed. We show that skin detection in video can be enhanced by exploiting the knowledge of the range of possible skin colours for the camera used. In normalized colour coordinates this range has a distinct shape we call the skin locus. We developed an adaptive histogram backprojection technique where the skin colour model is updated by pixels in the search region which fall in the skin locus. We demonstrate increased detection capability with webcam videos of faces taken successively under daylight, incandescent lamp, fluorescent light and a combination of these light sources
Article
Full-text available
This paper presents a study of three important issues of the color pixel classification approach to skin segmentation: color representation, color quantization, and classification algorithm. Our analysis of several representative color spaces using the Bayesian classifier with the histogram technique shows that skin segmentation based on color pixel classification is largely unaffected by the choice of the color space. However, segmentation performance degrades when only chrominance channels are used in classification. Furthermore, we find that color quantization can be as low as 64 bins per channel, although higher histogram sizes give better segmentation performance. The Bayesian classifier with the histogram technique and the multilayer perceptron classifier are found to perform better compared to other tested classifiers, including three piecewise linear classifiers, three unimodal Gaussian classifiers, and a Gaussian mixture classifier.
Conference Paper
Full-text available
The existence of large image datasets such as photos on the World Wide Web make it possible to build powerful generic models for low-level image attributes like color using simple histogram learning techniques. We describe the construction of color models for skin and non-skin classes from a dataset of nearly 1 billion labeled pixels. These classes exhibit a surprising degree of separability which we exploit by building a skin pixel detector that achieves an equal error rate of 88%. We compare the performance of histogram and mixture models in skin detection and find histogram models to be superior in accuracy and computational cost. Using aggregate features computed from the skin detector we build a remarkably effective detector for naked people. We believe this work is the most comprehensive and detailed exploration of skin color models to date
Article
Full-text available
The existence of large image datasets such as the set of photos on the World Wide Web make it possible to build powerful generic models for low-level image attributes like color using simple histogram learning techniques. We describe the construction of color models for skin and non-skin classes from a dataset of nearly 1 billion labelled pixels. These classes exhibit a surprising degree of separability which we exploit by building a skin pixel detector achieving a detection rate of 80% with 8.5% false positives. We compare the performance of histogram and mixture models in skin detection and find histogram models to be superior in accuracy and computational cost. Using aggregate features computed from the skin pixel detector we build a surprisingly effective detector for naked people. Our results suggest that color can be a more powerful cue for detecting people in unconstrained imagery than was previously suspected. We believe this work is the most comprehensive and detailed exploration of skin color models to date.
Article
Full-text available
This paper is concerned with estimating a probability density function of human skin color using a finite Gaussian mixture model whose parameters are estimated through the EM algorithm. Hawkins' statistical test on the normality and homoscedasticity (common covariance matrix) of the estimated Gaussian mixture models is performed and McLachlan's bootstrap method is used to test the number of components in a mixture. Experimental results show that the estimated Gaussian mixture model fits skin images from a large database. Applications of the estimated density function in image and video databases are presented. Keywords: Gaussian Mixture Model, EM algorithm, Skin Color, Skin Detection 1. INTRODUCTION Human skin color has been used and proved to be an effective feature in many applications from human face detection to hand tracking. However, most studies use either simple thresholding or a single Gaussian distribution to characterize the properties of skin color. Although skin colors o...
Article
In this paper we propose a simple and reliable approach for skin region segmentation to generate region-of-interest (ROI) for various human-computer interaction based applications. Performance of this approach is much better than existing approaches in terms of segmenting solid skin regions without generating much noisy segments. Moreover, it does not need any prior training session and it can adapt to detect skin pixels from images of people from different races taken at different imaging conditions.
Conference Paper
This study designs a novel skin-color extraction method to automatically select the ROI regions for videophone and videoconferencing applications. We propose a linear QP prediction at the macroblock layer to control the qualities at different regions for region-based H.263+ video codec with the option mode of modified quantization
Article
Skin detection plays an important role in a wide range of image processing applications ranging from face detection, face tracking, gesture analysis, content-based image retrieval systems and to various human computer interaction domains. Recently, skin detection methodologies based on skin-color information as a cue has gained much attention as skin-color provides computationally effective yet, robust information against rotations, scaling and partial occlusions. Skin detection using color information can be a challenging task as the skin appearance in images is affected by various factors such as illumination, background, camera characteristics, and ethnicity. Numerous techniques are presented in literature for skin detection using color. In this paper, we provide a critical up-to-date review of the various skin modeling and classification strategies based on color information in the visual spectrum. The review is divided into three different categories: first, we present the various color spaces used for skin modeling and detection. Second, we present different skin modeling and classification approaches. However, many of these works are limited in performance due to real-world conditions such as illumination and viewing conditions. To cope up with the rapidly changing illumination conditions, illumination adaptation techniques are applied along with skin-color detection. Third, we present various approaches that use skin-color constancy and dynamic adaptation techniques to improve the skin detection performance in dynamically changing illumination and environmental conditions. Wherever available, we also indicate the various factors under which the skin detection techniques perform well.
Article
In this paper, a novel algorithm for oriental face detection is presented to locate multiple faces in color scenery images. A binary skin color map is first obtained by applying the skin/non-skin color classification algorithm. Then, color regions corresponding to the facial and non-facial areas in the color map are separated with a clustering-based splitting algorithm. Thereafter, an elliptic face model is devised to crop the real human faces through the shape location procedure. Last, local thresholding technique and a statistic-based verification procedure are utilized to confirm the human faces. The proposed detection algorithm combines both the color and shape properties of faces. In this work, the color span of human face can be expanded as wilder as possible to cover different faces by using the clustering-based splitting algorithm. Experimental results reveal the feasibility of our proposed approach in solving face detection problem.
Conference Paper
We propose an adaptive skin-detection method, which allows modelling and detection of the true skin-color pixels with significantly higher accuracy and flexibility than previous methods. In principle, the proposed approach follows a two-step process. For a given image, we first perform a rough skin classification using a generic skin-model which defines the Skin-Similar space. The Skin-Similar space often contains many non-skin pixels due to the inevitable overlap in the color space between skin pixels and some non-skin pixels under the generic skin-model. The objective of the second step is to reduce the false-positive rate by analyzing the image under consideration. Specifically, in the second step, a Gaussian Mixture Model (GMM), specific to the image under consideration and refined from its Skin-Similar space, is derived using the standard Expectation-Maximization (EM) algorithm. We then use a Support Vector Machine (SVM) classifier to identify the skin Gaussian from the trained GMM by incorporating spatial and shape information of the skin pixels. Moreover, we examine how the improvement on skin detection by this adaptive skin-model impacts the detection accuracy in the application of Objectionable Image Filtering. We further propose a two-level classification scheme based on hierarchical bagging to improve the accuracy. Results of extensive experiments on large databases demonstrate the effectiveness and benefits of our adaptive skin-model.
Conference Paper
We propose a new image classification technique that utilizes neural networks to classify skin and non-skin pixels in color images. The aim is to develop a universal and robust model of the human skin color that caters for all human races. The ability to detecting solid skin regions in color images by the model is extremely useful in applications such as face detection and recognition, and human gesture analysis. Experimental results show that the neural network classifiers can consistently achieve up to 90% accuracy in skin color detection
Article
Image segmentation is the process by which an original image is partitioned into some homogeneous regions. In this paper, a novel multiresolution color image segmentation (MCIS) algorithm which uses Markov random fields (MRF's) is proposed. The proposed approach is a relaxation process that converges to the MAP (maximum a posteriori) estimate of the segmentation. The quadtree structure is used to implement the multiresolution framework, and the simulated annealing technique is employed to control the splitting and merging of nodes so as to minimize an energy function and therefore, maximize the MAP estimate. The multiresolution scheme enables the use of different dissimilarity measures at different resolution levels. Consequently, the proposed algorithm is noise resistant. Since the global clustering information of the image is required in the proposed approach, the scale space filter (SSF) is employed as the first step. The multiresolution approach is used to refine the segmentation. Experimental results of both the synthesized and real images are very encouraging. In order to evaluate experimental results of both synthesized images and real images quantitatively, a new evaluation criterion is proposed and developed
Article
The ability to give higher priority to regions-of-interest (ROI) is the emerging functionality for present day video coding. A simple and fast method for face detection is proposed to define ROIs dynamically in real time applications. We use the color information Cr and RGB variance to determine the skin-color pixels. Because the two color spaces are commonly used in most hardware and video codec standards, there is no extra computation overhead required for conversion. Low-pass filtering is applied to the background to reduce used bits. For the video coding system, a region-based video codec based on H.263+ with the option mode of modified quantization is set up. We adjust the distortion weight parameter and variance at the macroblock layer to control the qualities at different regions. Experimental results show that the proposed methods can significantly improve quality at the ROI. Our methods are suitable for real time videoconferencing.
Article
Skin color has proven to be a useful and robust cue for face detection, localization and tracking. Image content filtering, contentaware video compression and image color balancing applications can also benefit from automatic detection of skin in images. Numerous techniques for skin color modelling and recognition have been proposed during several past years. A few papers comparing different approaches have been published [Zarit et al. 1999], [Terrillon et al. 2000], [Brand and Mason 2000]. However, a comprehensive survey on the topic is still missing. We try to fill this vacuum by reviewing most widely used methods and techniques and collecting their numerical evaluation results.