• 한국정보전자통신기술학회논문지
• /
• 제6권4호
• /
• pp.249-254
• /
• 2013

In the area of computer vision, face recognition is being intensively researched. It is generally known that before a face is recognized it must be localized. Skin-color information is an important feature to segment skin-color regions. To extract skin-color regions the skin-color model based on multi-layer perceptron has been proposed. Extracted regions are analyzed to emphasize ellipsoidal regions. The results from this study show good accuracy for our vehicle driver face detection system.

• 한국멀티미디어학회논문지
• /
• 제25권9호
• /
• pp.1367-1374
• /
• 2022

In this paper, we propose an automatic cosmetic foundation recommendation system that suggests a good foundation product based on the user's skin color. The proposed system receives and preprocesses user images and detects skin color with OpenCV and machine learning algorithms. The system then compares the performance of the training model using XGBoost, Gradient Boost, Random Forest, and Adaptive Boost (AdaBoost), based on 550 datasets collected as essential bestsellers in the United States. Based on the comparison results, this paper implements a recommendation system using the highest performing machine learning model. As a result of the experiment, our system can effectively recommend a suitable skin color foundation. Thus, our system model is 98% accurate. Furthermore, our system can reduce the selection trials of foundations against the user's skin color. It can also save time in selecting foundations.

• 한국통신학회논문지
• /
• 제36권10C호
• /
• pp.621-628
• /
• 2011

영상에서 피부색은 얼굴 영상 기반의 검출 및 인식 시스템에서 중요한 단서로 사용되지만 영상 획득시의 광원에 따라 상이하게 표현되기 때문에 사용의 어려움이 있다. 이러한 문제를 위해 본 논문에서는 Retinex, Gray World, White Patch, Simplified Horn, Shades of grey, 그리고 Edge-Based color constancy 색 항등성 기법을 대상으로 피부색 복원에 관한 성능을 비교하고, 자연 영상을 이용하여 피부색 검출 성능을 평가한다. 이를 위해 Caltech Face Database의 영상들에 유색 광원의 효과를 부가하여 생성된 실험 영상을 생성했다. 이후 각 기법에 따른 피부색 복원의 일관성을 수치적으로 평가하기 위하여 Cb-Cr 히스토그램을 기반으로 결과 영상들의 표준편차를 계산하였으며, 피부색 검출 성능 실험을 위해서 YCbCr과 RGB 기반의 경출 기법을 사용하였다. 실험 결과 Gray World 기법은 타 기법에 비해 높은 성능을 나타냈으며, 색 항등성 기법의 적용을 통해 광원의 영향을 받은 피부색을 색 공간내의 일정한 범위로 복원하는 것이 가능하였다.

• 한국멀티미디어학회:학술대회논문집
• /
• 한국멀티미디어학회 2002년도 추계학술발표논문집
• /
• pp.85-88
• /
• 2002

얼굴 영역을 검출하는 데 있어서 가장 기본적이면서도 중요한 정보가 컬러 정보이다. 하지만 컬러정보는 사용하는 컬러모델링 및 얼굴의 Skin Color를 평가하는 범위를 어떻게 정의하느냐에 따라 얼굴의 검출 성능에 많은 영향을 끼친다. 본 논문에서는 얼굴 영역을 검출하기 위한 첫 번째 조건으로 Skin color영역을 색상값과 다양한 데이터로부터 명암도에 따른 Skin color의 분포와 비율을 학습 함으로써 Skin color 영역을 검출 성능을 높이며, 퍼지 아트 알고리즘을 이용하여 얼굴과 비얼굴 데이터에 인증함으로써 얼굴 영역의 검출 성능을 높인다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.49.1-49
• /
• 2001

The skin color model is a very important concept in face detection, face recognition and face tracking. Usually, this model is obtained by estimating a probability density function of skin color distribution. In many cases, it is assumed that the underlying density function follows a Gaussian distribution. In this paper, a new method for non-parametric estimation of the probability density function, by using feed-forward neural network, is used to estimate the underlying skin color model. By using this method, the resulting skin color model is better than the Gaussian estimation and substantially approaches the real distribution. Applications to face detection and face ...

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 하계종합학술대회 논문집(4)
• /
• pp.49-52
• /
• 2002

This paper presents a fast algorithm for face detection from color images on internet. We use Mahalanobis distance between standard skin color and actual pixel color on IQ color space to segment skin color regions. The skin color regions are the candidate face region. Further, the locations of eyes and mouth regions are found by computing average pixel values on horizontal and vertical pixel lines. The geometry of mouth and eye locations is compared to the standard face geometry to eliminate false face regions. Our Method is simple and fast so that it can be applied to face search engine for internet.

• 한국의류학회지
• /
• 제26권2호
• /
• pp.303-314
• /
• 2002

The colors of apparel have become an important element to be used strategically in order to give differentiated character at the level of fiber and fabric production. The colors of apparel have a close relationship with the skin colors of consumers and their preference colors. This study was carried out to classify the skin colors of Korean elderly women into several similar skin colors and to analyse their preference colors. Sample size was 471 Korean elderly women. With color spectrometer, JX-777, we measured 4 points of the body; cheek with removing cosmetics off, forehead, rear neck and arm on the interior part near elbow. All subjects had been shown with 40 color chips and answered the preference colors of apparel and the preference colors. Data weirs analysed to classify skin colors using K-means Cluster Analysis and Duncan test. Independent variables for Cluster Analysis were 12 variables out of L value, a value and b value of 4 points. In doing so, we used SPSS WIN 10 statistical package. Findings were as follows: 1) The skin colors of the Korean elderly women were composed of skin colors of YR, R, and Y. 2) 355 subjects were classified into 4 kinds of skin color groups. 3) The average face color of type 1 was 6.7YR 5.1/4.3 and 56 observations out of 355 subjects were composed of Type 1 and of Type 2 was 6.1YR 6.1/4.5 and 166 observations out of 355 and of 3 Type 6. YR 4.8/4.2 and 75 observations out of 355 and of Type was 6.17 YR 5.7/4.7 and 58 observations out of 355. 4) The average skin color of Type 1 was 7.0YR 5.9/4.4 and of Type 2 was 7.2YR 6.3/4.2 and of Type 3 was 7.0YR 6.2/4.2 and of Type 4 was 7.6YR 5.4/4.2 respectively. 5) The mean values of 12 variables between the 4 classified face color and skin color groups showed significantly different except H value of skin color. 6) All 4 groups showed that the most preference color of apparel and the most preference color were 2.5R 5/14 respectively.

• 한국컴퓨터정보학회논문지
• /
• 제15권12호
• /
• pp.27-35
• /
• 2010

유해영상 검출에서 정확하게 피부 색상 영역을 획득하는 것은 매우 중요하다. 그러나 기존의 방법들은 서로 다른 인종, 조명, 화장, 사용된 카메라 등과 같은 여러 원인으로 인해 피부 색상 추출에 여전히 문제를 가지고 있으며, 사전에 미리 정해진 피부 색상 분포 모델을 이용하여 영상에서 피부 영역을 검출한다. 이러한 문제를 해결하기 위해 본 논문에서는 눈 주변 영역에서 샘플을 추출하여 입력 영상에 최적으로 적합된 피부 색상 분포 모델을 생성하여 피부 영역을 강건하게 분할하고, 분할된 피부 영역에서 성인 영상을 대표할 수 있는 특징을 추출한 후, 분할된 피부 영역이 나체의 몸체를 포함하고 있는지를 뉴럴 네트워크 다층 퍼셉트론을 통해 여러 대표적인 특징들을 통합하면서 추론하는 새로운 방법을 제안하다. 본 논문의 실험에서는 피부 색상 영역 분할과 성인영상 검출의 두 가지 성능 측면에서 제안한 방법의 성능이 기존의 방법에 비해 보다 우수함을 보인다. 본 논문에서 제안한 강건한 유해영상검출 기법은 얼굴 검출, 성인영상 필터링 등과 같은 관련된 여러 응용 분야에서 유용하게 활용될 것이라 기대된다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.579-582
• /
• 1999

In this paper, we propose a human face detection algorithm using adaptive skin color model and neural networks. To attain robustness in the changes of illumination and variability of human skin color, we perform a color segmentation of input image by thresholding adaptively in modified hue-saturation color space (TSV). In order to distinguish faces from other segmented objects, we calculate invariant moments for each face candidate and use the multilayer perceptron neural network of backpropagation algorithm. The simulation results show superior performance for a variety of poses and relatively complex backgrounds, when compared to other existing algorithm.

• 한국멀티미디어학회논문지
• /
• 제18권5호
• /
• pp.610-619
• /
• 2015

*In this paper, we propose a real-time face tracking system using an adaptive face detector and a tracking algorithm. An image is divided into the regions of background and face candidate by a real-time updated skin color identifying system in order to accurately detect facial features. The facial characteristics are extracted using the five types of simple Haar-like features. The extracted features are reinterpreted by Principal Component Analysis (PCA), and the interpreted principal components are processed by Support Vector Machine (SVM) that classifies into facial and non-facial areas. The movement of the face is traced by Kalman filter and Mean shift, which use the static information of the detected faces and the differences between previous and current frames. The proposed system identifies the initial skin color and updates it through a real-time color detecting system. A similar background color can be removed by updating the skin color. Also, the performance increases up to 20% when the background color is reduced in comparison to extracting features from the entire region. The increased detection rate and speed are acquired by the usage of Kalman filter and Mean shift.