• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2013년도 춘계학술대회
• /
• pp.364-367
• /
• 2013

본 논문은 모폴로지 연산을 이용하여 얼굴의 잡티를 제거하는 기법을 제안한다. 먼저, 입력받은 RGB 3채널 영상에서 각 채널의 픽셀 값을 이용하여 피부 영역을 검출한다. 피부 영역의 R, G, B 각 채널 영상에 대한 히스토그램을 만들어 각 채널 당 빈도수가 많은 픽셀 값 3개를 저장한다. 그 후, 피부 영역에 블랙햇 연산을 하여 잡티를 찾아낸다. 잡티로 검출된 픽셀은 자신의 픽셀 값과 8방향 이웃 픽셀 값, 앞에서 선택된 픽셀 3개의 평균으로 그 값을 변경한다. 그리고 잡티 부분의 픽셀에 중간값 필터를 이용하여 블러링을 한다. 잡티가 있는 얼굴 사진을 테스트 영상으로 실험한 결과 제안된 시스템은 얼굴의 잡티를 제거하여 얼굴 피부를 보정하는 방법은 단순히 영상을 밝게 하여 얼굴 피부를 보정하는 방법보다 효율적이다.

• 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.

• 한국콘텐츠학회논문지
• /
• 제9권5호
• /
• pp.1-12
• /
• 2009

올바른 피부색 검출은 사람의 얼굴 검출 및 동작 분석에서 매우 중요한 전처리과정에 속한다. 피부 검출은 일반적으로 화소의 칼라 공간을 Non-RGB로 변형하고, 피부색의 조명 요소를 제거한 다음 피부색 분포 모델에 의해 Skin과 Non-Skin으로 분류하는 3단계로 진행된다. 이는 피부색 검출이 칼라 공간, 조명 요소의 존재 여부, 피부 모델링 방법에 따라 수행 성능에 많은 영향을 받기 때문이다. 본 연구에서는 조명 조건에 따라 피부색 모델의 범위에 차이가 있다는 사실에 기초하여 다양한 조명 조건과 복잡한 배경을 가진 영상에서 효과적으로 다인종의 피부색을 분류해내 기 위한 3차원 피부색 모델을 제시하고자 한다. 제안된 피부색 모델은 화소의 칼라 공간을 YCbCr공간으로 변형하고, 각 요소(Y, Cb, Cr) 값에 의한 3차원 피부색 모델을 형성한다. 다인종의 피부색을 함께 분할하기 위해 인종(백인, 흑인, 황인)별 피부색 모델을 먼저 생성한 후 각각의 모델에서 피부색 확률에 따라 결합한 다인종을 위한 통합 모델을 생성하였다. 또한 우리는 적은 양의 훈련 데이터로 피부색 영역을 올바르게 검출할 수 있도록 여러 단계의 피부색 영역을 설정하였다.

• Journal of Information Technology Applications and Management
• /
• 제12권1호
• /
• pp.231-239
• /
• 2005

This paper describes a new algorithm for pixel-based skin color detection to differentiate human form in color images by the ratio of R to H. In order to detect skin color efficiently, we examine the distribution of the R, G and B color elements combining to constitute the skin color in various color images. It shows that R is located in a narrower area than G and B on the RGB color space. And skin color is more related to R than G and B. Meanwhile, when the color image is transformed to the HSI color space, the S is variously changed in accordance with skin colors. The I is changed in accordance with the quantity and angle of light. But the H is less influenced by other conditions except for color. On the basis of the aforementioned study, we propose that the threshold for skin color detection is decided by the ratio of R to H. The proposed method narrows down the range of threshold, detects more skin color and reduces mis-detection of skin color in comparison to detection by R or H. In experimentation. it shows that the proposed algorithm overcomes changes of brightness and color to detect skin color in color images.

• ETRI Journal
• /
• 제28권6호
• /
• pp.770-776
• /
• 2006

This paper presents a hyperspectral imaging technique based on laser-induced fluorescence for non-invasive detection of tumorous tissue on mouse skin. Hyperspectral imaging sensors collect image data in a number of narrow, adjacent spectral bands. Such high-resolution measurement of spectral information reveals contiguous emission spectra at each image pixel useful for the characterization of constituent materials. The hyperspectral image data used in this study are fluorescence images of mouse skin consisting of 21 spectral bands in the visible spectrum of the wavelengths ranging from 440 nm to 640 nm. Fluorescence signal is measured with the use of laser excitation at 337 nm. An acousto-optic tunable filter (AOTF) is used to capture images at 10 nm intervals. All spectral band images are spatially registered with the reference band image at 490 nm to obtain exact pixel correspondences by compensating the spatial offsets caused by the refraction differences in AOTF at different wavelengths during the image capture procedure. The unique fluorescence spectral signatures demonstrate a good separation to differentiate malignant tumors from normal tissues for rapid detection of skin cancers without biopsy.

• 한국정보통신학회논문지
• /
• 제14권8호
• /
• pp.1809-1818
• /
• 2010

피부색 영역의 검출을 위한 기존 연구들은 영상의 각 픽셀을 피부에 속하는 픽셀(피부픽셀)과 속하지 않는 픽셀(비피부픽셀)로 나누게 된다. 이때 정확한 피부색 영역을 검출하는 작업은 영상의 조명효과 및 화장에 의한 피부색 변형 등으로 매우 어려운 작업이다. 본 논문에서는 피부 영역 검출을 어렵게 하는 여러 가지 요인을 포함한 영상들로부터 효율적으로 피부영역을 검출하기 위해 계층화된 피부 모델과 컨텍스트 정보를 통합하여 피부 영역 검출의 성능을 향상시키는 방법을 제안한다. 먼저, 획득된 영상들로부터 뽑아낸 피부색 색깔 값들의 확률분포를 YCbCr칼라 공간에 만들고, 그 확률 값에 따라 피부(Skin), 피부후보(Skinness), 비피부(Non-skin)의 3계층으로 분류한 3차원 피부색 모델을 만든다. 계층화된 피부색 모델을 이용하여 각 픽셀의 피부색 여부를 결정하고, 피부후보(Skinness)색에 해당하는 경우에는 이웃 화소의 정보를 고려하여 피부색 또는 비 피부색으로 정하게 된다. 제안 방법의 사용으로 피부색이 왜곡 되었거나 피부색과 유사한 객체가 포함된 다양한 영상들에서도 효율적으로 피부 영역을 분할할 수 있었다.

• 한국정보기술학회 영문논문지
• /
• 제8권2호
• /
• pp.47-58
• /
• 2018

In this paper, we propose a method for skin condition analysis using a camera module embedded in a smartphone without a separate skin diagnosis device. The type of skin disease detected in facial image taken by smartphone is acne, pigmentation, blemish and flush. Face features and regions were detected using Haar features, and skin regions were detected using YCbCr and HSV color models. Acne and flush were extracted by setting the range of a component image hue, and pigmentation was calculated by calculating the factor between the minimum and maximum value of the corresponding skin pixel in the component image R. Blemish was detected on the basis of adaptive thresholds in gray scale level images. As a result of the experiment, the proposed skin condition analysis showed that skin diseases of acne, pigmentation, blemish and flush were effectively detected.

• 비파괴검사학회지
• /
• 제30권3호
• /
• pp.181-193
• /
• 2010

The paper presents an alternative way to classical stereocorrelation. First, 2D image processing of random patterns is described. Sub-pixel displacements are determined using phase analysis. Then distortion evaluation is presented. The distortion is identified without any assumption on the lens model because of the use of a grid technique approach. Last, shape measurement and shape variation is caught by fringe projection. Analysis is based on two pin-hole assumptions for the video-projector and the camera. Then, fringe projection is coupled to in-plane displacement to give rise to 3D measurement set-up. Metrological characterization shows a resolution comparable to classical (stereo) correlation technique ($1/100^{th}$ pixel). Spatial resolution seems to be an advantage of the method, because of the use of temporal phase stepping (shape measurement, 1 pixel) and windowed Fourier transform (in plane displacements measurement, 9 pixels). Two examples are given. First one is the study of skin properties; second one is a study on leather fabric. In both cases, results are convincing, and have been exploited to give mechanical interpretation.

• 한국정보기술응용학회:학술대회논문집
• /
• 한국정보기술응용학회 2005년도 6th 2005 International Conference on Computers, Communications and System
• /
• pp.153-156
• /
• 2005

Skin region detection in images is an important process in many computer vision applications targeting humans such as hand gesture recognition and face identification. It usually starts at a pixel-level, and involves a pre-process of color spae transformation followed by a classification process. A color space transformation is assumed to increase separability between skin classes and other classes, to increase similarity among different skin tones, and to bring a robust performance under varying imaging conditions, without any complicated analysis. In this paper, we examine if the color space transformation actually brings those benefits to the problem of skin region detection on a set of human hand images with different postures, backgrounds, people, and illuminations. Our experimental results indicate that color space transfomation affects the skin detection performance. Although the performance depends on camera and surround conditions, normalized [R, G, B] color space may be a good choice in general.

• Journal of Electrical Engineering and Technology
• /
• 제12권5호
• /
• pp.2067-2078
• /
• 2017

In this paper, a new face tracking algorithm is proposed. The CamShift (Continuously adaptive mean SHIFT) algorithm shows unstable tracking when there exist objects with similar color to that of face in the background. This drawback of the CamShift is resolved by the proposed algorithm using Kinect's pixel-by-pixel depth information and the skin detection method to extract candidate skin regions in HSV color space. Additionally, even when the target face is disappeared, or occluded, the proposed algorithm makes it robust to this occlusion by the feature point matching. Through experimental results, it is shown that the proposed algorithm is superior in tracking performance to that of existing TLD (Tracking-Learning-Detection) algorithm, and offers faster processing speed. Also, it overcomes all the existing shortfalls of CamShift with almost comparable processing time.