• 대한한의학회지
• /
• 제44권4호
• /
• pp.136-149
• /
• 2023

Objectives: Facial diagnosis is an important part of clinical diagnosis in traditional East Asian Medicine. In this paper, we proposed a model to quantitatively classify cold sensitivity using a fully automated facial image analysis system. Methods: We investigated cold sensitivity in 452 subjects. Cold sensitivity was determined using a questionnaire and the Cold Pattern Score (CPS) was used for analysis. Subjects with a CPS score below the first quartile (low CPS group) belonged to the cold non-sensitivity group, and subjects with a CPS score above the third quartile (high CPS group) belonged to the cold sensitivity group. After splitting the facial images into train/validation/test sets, the train and validation set were input into a convolutional neural network to learn the model, and then the classification accuracy was calculated for the test set. Results: The classification accuracy of the low CPS group and high CPS group using facial images in all subjects was 76.17%. The classification accuracy by sex was 69.91% for female and 62.86% for male. It is presumed that the deep learning model used facial color or facial shape to classify the low CPS group and the high CPS group, but it is difficult to specifically determine which feature was more important. Conclusions: The experimental results of this study showed that the low CPS group and the high CPS group can be classified with a modest level of accuracy using only facial images. There was a need to develop more advanced models to increase classification accuracy.

• Imaging Science in Dentistry
• /
• 제51권3호
• /
• pp.279-290
• /
• 2021

Purpose: Using images in the facial image comparison process poses a challenge for forensic experts due to limitations such as the presence of facial expressions. The aims of this study were to analyze how morphometric changes in the face during a spontaneous smile influence the facial image comparison process and to evaluate the reproducibility of measurements obtained by digital stereophotogrammetry in these situations. Materials and Methods: Three examiners used digital stereophotogrammetry to obtain 3-dimensional images of the faces of 10 female participants(aged between 23 and 45 years). Photographs of the participants' faces were captured with their faces at rest (group 1) and with a spontaneous smile (group 2), resulting in a total of 60 3-dimensional images. The digital stereophotogrammetry device obtained the images with a 3.5-ms capture time, which prevented undesirable movements of the participants. Linear measurements between facial landmarks were made, in units of millimeters, and the data were subjected to multivariate and univariate statistical analyses using Pirouette^® version 4.5 (InfoMetrix Inc., Woodinville, WA, USA) and Microsoft Excel^® (Microsoft Corp., Redmond, WA, USA), respectively. Results: The measurements that most strongly influenced the separation of the groups were related to the labial/buccal region. In general, the data showed low standard deviations, which differed by less than 10% from the measured mean values, demonstrating that the digital stereophotogrammetry technique was reproducible. Conclusion: The impact of spontaneous smiles on the facial image comparison process should be considered, and digital stereophotogrammetry provided good reproducibility.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제54권9호
• /
• pp.577-579
• /
• 2005

In this study, the automatic tracking algorithm tracing a human face is proposed by using YCbCr color coordinated information and its thermal properties expressed in terms of thermal indexes in an infra-red image. The facial candidates are separately estimated in CbCr color and infra-red domain, respectively with applying the morphological image processing operations and the geometrical shape measures for fitting the elliptical features of a human face. The identification of a true face is accomplished by logical 'AND' operation between the refined image in CbCr color and infra-red domain.

• 지능정보연구
• /
• 제29권2호
• /
• pp.171-187
• /
• 2023

저화소의 감시카메라와 같은 촬영 장비를 통해 사람의 얼굴을 인식할 경우, 화질이 낮아 얼굴을 포착하기 어렵다는 문제점이 있다. 이렇게, 사람의 얼굴을 인식하기 어렵다면 범죄용의자나 실종자를 특정해내지 못하는 등의 문제가 발생할 수 있다. 기존 이미지 속 안면 인식에 관한 연구들에서는 정제된 데이터셋을 사용하였기 때문에 다양한 환경에서의 성능을 가늠하기 어렵다는 한계가 존재한다. 이에, 본 논문에서는 저화질 이미지에서 안면 인식 성능이 떨어지는 문제를 해결하기 위해 다양한 환경을 고려한 저화질 안면 이미지에 대해 화질 개선을 수행하여 고화질 이미지를 생성한 뒤, 안면 특징점 검출의 성능 향상시키는 방법을 제안한다. 제안 방법의 현실 적용 가능성을 확인하기 위해 전체 이미지에서 사람이 상대적으로 작게 나타나는 데이터셋을 선정하여 실험을 수행하였다. 또한 마스크 착용 상황을 고려한 안면 이미지 데이터셋을 선정하여, 현실 문제로의 확장 가능성을 탐구하였다. 안면 이미지의 화질을 개선하여 특징점 검출 모델의 성능을 측정한 결과, 개선 후 안면의 검출 여부는 마스크를 착용하지 않은 이미지의 경우 평균 3.47배, 마스크를 착용한 경우 평균 9.92배로 성능 향상을 확인할 수 있었다. 안면 특징점에 대한 RMSE는 마스크를 착용한 이미지의 경우 평균 8.49배 감소, 마스크를 착용하지 않은 경우 평균 2.02배 감소한 것을 확인할 수 있었다. 이에, 화질 개선을 통해 저화질로 포착된 안면 이미지에 대한 인식률을 높여 제안 방법의 활용 가능성을 확인할 수 있었다.

• 패션비즈니스
• /
• 제19권5호
• /
• pp.34-47
• /
• 2015

Sexiness is the image of a person who attracts attention regardless of his/her age, and today's society has created a more positive impression by transforming the image into something others consider even more attractive. Consequently, we begin to take interest in makeup, - a means of portraying a good impression. Eyebrow makeup in particular, being at the center of the determining factor of a good facial image, can be thought of as the makeup that can most effectively transform one's image. The purpose of this study was to analyze the difference in perceived sexiness depending on the general perception of eyebrow makeup. This study produced an eyebrow stimulus that applied the average rate of change in an image transformation of different eyebrows in order to raise the objective credibility of the sensibility evaluation so it could determine the figure of influence that eyebrows had on facial impressions. The research results showed that the majority of female university students believed that eyebrows were an effective means of expression in changing facial images and attributed a higher mature and sexy image if the average ratio of change was higher. The study, could also identify that a sexy image was perceived when the average rate of change was between 0.39~0.44. In addition, when the gradient of the shape of the viewers' own eyebrows was low, it could be verified that they perceived an image to be sexy from seeing eyebrows with a relatively high gradient.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제36권7호
• /
• pp.548-559
• /
• 2009

Viola와 Jones가 제안한 에이다부스트 얼굴 검출기는 속도와 정확도면에서 매우 훌륭한 성능을 보이고 있는 얼굴 검출기이다. 하지만 에이다부스트 얼굴 검출기에도 여전히 오검출이 발생하며, 이를 줄이기 위해서는 더욱 많은 계산이 필요하다. 에이다부스트 얼굴 검출기는 흑백 영상만을 사용하므로, 색상정보를 사용하면 더 적은 연산으로 오검출율을 낮출 수 있다. 본 논문은 얼굴 색상 정보를 이용하여 대상 영상에서 부 윈도우를 효율적으로 검색하고, 에이다부스트 얼굴 검출기의 첫 단계에 계산속도가 매우 빠른 얼굴 색상을 이용한 얼굴/비얼굴 분류기를 채용하여 더 빠른 얼굴 검출 속도와 더 낮은 오검출율을 달성할 수 있는 단계별 얼굴 검출 모텔을 제안하였다. 얼굴색상 필터링을 위해 정의된 얼굴색상 소속함수를 이용하여 얼굴색상 필터 영상과 그 누적영상을 계산한다. 누적 영상에 의해 빠른 속도로 임의의 부 윈도우의 밀도를 계산할 수 있다. 제안된 검색 방법은 이 색상 밀도에 기반하여 얼굴일 가능성이 없는 부 윈도우들을 생략하게 된다. 그리고 부 윈도우의 밀도를 이용한 얼굴/비얼굴 분류기는 단계별 얼굴 검출기의 앞단에서 얼굴이 아닌 부 윈도우를 빠르게 거절한다. 제안된 얼굴 검출 모델은 적은 계산으로 오검출율을 낮출 수 있었으며, 실시간 얼굴 검출 속도를 비약적으로 향상시킬 수 있었다.

• 한국정보처리학회논문지
• /
• 제7권2호
• /
• pp.569-576
• /
• 2000

일반화 대칭 변환(Generalized Symmetry Transform : GST)은 디칭성을 이용하여 영상의 사전 정보 없이 얼굴 특징의 위치를 추출할 수 있는 방법이다. 그러나, 눈, 코, 입 등의 특징보다 마스크의 크기가 커야 하므로 많은 처리시간이 필요하다. 그리고, 얼굴의 특징을 결정하기 위해 통계적 처리가 수반되는 중심선 계산으로 인해 처리 과정이 복잡하다. 이러한 단점을 개선하기 위해, 본 논문에서는 두 가지 방법을 제안하였다. 첫째, 처리 시간을 단축하기 위해 원 영상 대신 영상의 정보를 충분히 가지는 축소 영상을 사용하였다. 둘째, 중심선 계산을 위해 복잡한 통계적 처리 대신 추출된 첨두치의 위치를 이용하였다. 제안된 방법의 성능을 살펴보기 위해, 정면, 회전, 안경, 수염이 있는 영상 등 200개의 영상에 대해 실험하였다. 그 결과, 제안된 방법은 85%의 특징 탐지율과 기존의 방법에 비해 약 53배 이상감소된 처리시간을 나타내었다.

• 패션비즈니스
• /
• 제13권3호
• /
• pp.12-24
• /
• 2009

This paper aims to investigate if a total of 266 college students from Yanbian, China, and Korea's Jeonnam Province are satisfied with their face and body type, and to find out an image-development method, depending on body characteristics. For this, a questionnaire survey using the 5-point Likert Scale was conducted, and an SPSS program has been used for data analysis. Besides the descriptive analysis, crosstab analysis, t-verification, and frequency analysis have been conducted. As body shape, posture, costume, accessories and makeup play an important role in image formation, this paper intends to form positive ego through exact awareness of the body shape by providing base data to a set image-development strategy. Then, the result has turned out as follows: First, in terms of the body index (i.e., Rohrer index), both Korea's Jeonnam Province and China's Yanbian college students were included in a category of the mean value. In terms of satisfaction with their body shape, on the contrary, China's China's Yanbian college students were higher than Korea's Jeonnam Province college students. Second, male China's Yanbian college students were slightly higher than female China's Yanbian college students in terms of satisfaction with their facial shape. However, the male students showed no big interest in facial care. Third, in terms of facial shape, an egg-shaped face was the most preferred in both China's Yanbian and Korea's Jeonnam Province college students, followed by an inverted triangle-shaped face in China's Yanbian college students and a diamond-shaped face in Korea's Jeonnam Province college students, showing significant difference ($p{\le}.001$). Even though both college students live in Northeast Asia, their preference on facial shape turned out to greatly differ, which indicates their different social environments. This paper will be helpful in global marketing for college students who are the major consumers in the future as Korean-Chinese exchange increases.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 학술대회 논문집 정보 및 제어부문
• /
• pp.147-149
• /
• 2004

There are no authentic solutions in a face region extraction problem though it is an important part of pattern recognition and has diverse application fields. It is not easy to develop the facial region extraction algorithm because the facial image is very sensitive according to age, sex, and illumination. In this paper, to solve these difficulties, a fuzzy color filer based on the facial region extraction algorithm is proposed. The fuzzy color filter makes the robust facial region extraction enable by modeling the skin color. Especially, it is robust in facial region extraction with various illuminations. In addition, to identify the fuzzy color filter, a linear matrix inequality(LMI) optimization method is used. Finally, the simulation result is given to confirm the superiority of the proposed algorithm.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 2000년도 춘계 학술대회 및 국제 감성공학 심포지움 논문집 Proceeding of the 2000 Spring Conference of KOSES and International Sensibility Ergonomics Symposium
• /
• pp.126-132
• /
• 2000

This paper presents a facial expression recognition based on Gabor wavelets that uses a fuzzy C-means(FCM) clustering algorithm and neural network. Features of facial expressions are extracted to two steps. In the first step, Gabor wavelet representation can provide edges extraction of major face components using the average value of the image's 2-D Gabor wavelet coefficient histogram. In the next step, we extract sparse features of facial expressions from the extracted edge information using FCM clustering algorithm. The result of facial expression recognition is compared with dimensional values of internal stated derived from semantic ratings of words related to emotion. The dimensional model can recognize not only six facial expressions related to Ekman's basic emotions, but also expressions of various internal states.