• 한국컴퓨터정보학회논문지
• /
• 제25권5호
• /
• pp.1-9
• /
• 2020

본 논문에서는 머리카락과 모자 영역의 마스크 정보를 활용하여 더 자연스러운 얼굴 속성 편집(facial attribute editing)을 수행하는 모델을 제안한다. 최신 얼굴 속성 편집 연구인 STGAN은 다중 얼굴 속성을 자연스럽게 편집하는 성과를 보였다. 그러나 머리카락과 관련된 속성을 편집할 때 부자연스러운 결과를 생성할 수 있다. 제안하는 방법의 핵심 아이디어는 기존 모델에서 부족했던 얼굴 영역의 정보를 모델에 추가로 반영하는 것이다. 이를 위해 세 가지 아이디어를 적용한다. 첫째로 마스크를 통해 머리카락 면적 속성을 추가하여 머리카락 정보를 보완한다. 둘째로 순환 일관성 손실(cycle consistency loss)을 추가하여 영상의 불필요한 변화를 억제한다. 셋째로 모자 분할 신경망을 추가하여 모자 영역 왜곡을 방지한다. 정성적 평가를 통해 제안하는 방법 적용 여부에 따른 유효성을 평가 및 분석한다. 실험 결과에서 제안하는 방법이 머리카락 및 얼굴 영역을 더 자연스럽게 생성하고, 모자 영역의 왜곡을 성공적으로 방지했다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -2
• /
• pp.1252-1255
• /
• 2002

In this paper, face recognition algorithm, by using skin color information of HSI color coordinate collected from face images, elliptical mask, fratures of face including eyes, nose and mouth, and geometrical feature vectors of face and facial angles, is proposed. The proposed algorithm improved face region extraction efficacy by using HSI information relatively similar to human's visual system along with color tone information about skin colors of face, elliptical mask and intensity information. Moreover, it improved face recognition efficacy with using feature information of eyes, nose and mouth, and Θ1(ACRED), Θ2(AMRED) and Θ 3(ANRED), which are geometrical face angles of face. In the proposed algorithm, it enables exact face reading by using color tone information, elliptical mask, brightness information and structural characteristic angle together, not like using only brightness information in existing algorithm. Moreover, it uses structural related value of characteristics and certain vectors together for the recognition method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권11호
• /
• pp.5605-5623
• /
• 2016

We propose a face recognition method that utilizes the LCP face descriptor. The proposed method applies a LoG mask to extract a face contour response, and employs the LCP algorithm to produce a binary pattern representation that ensures high recognition performance even under the changes in illumination, noise, and aging. The proposed LCP algorithm produces excellent noise reduction and efficiency in removing unnecessary information from the face by extracting a face contour response using the LoG mask, whose behavior is similar to the human eye. Majority of reported algorithms search for face contour response information. On the other hand, our proposed LCP algorithm produces results expressing major facial information by applying the threshold to the search area with only 8 bits. However, the LCP algorithm produces results that express major facial information with only 8-bits by applying a threshold value to the search area. Therefore, compared to previous approaches, the LCP algorithm maintains a consistent accuracy under varying circumstances, and produces a high face recognition rate with a relatively small feature vector. The test results indicate that the LCP algorithm produces a higher facial recognition rate than the rate of human visual's recognition capability, and outperforms the existing methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권10호
• /
• pp.3685-3707
• /
• 2021

In recent years, facial attribute editing has been successfully used to effectively change face images of various attributes based on generative adversarial networks and encoder-decoder models. However, existing models have a limitation in that they may change an unintended part in the process of changing an attribute or may generate an unnatural result. In this paper, we propose a model that improves the learning of the attention mask by adding a spatial attention mechanism based on the unified selective transfer network (referred to as STGAN) using semi-supervised learning. The proposed model can edit multiple attributes while preserving details independent of the attributes being edited. This study makes two main contributions to the literature. First, we propose an encoder-decoder model structure that learns and edits multiple facial attributes and suppresses distortion using an attention mask. Second, we define guide masks and propose a method and an objective function that use the guide masks for multiple facial attribute editing through semi-supervised learning. Through qualitative and quantitative evaluations of the experimental results, the proposed method was proven to yield improved results that preserve the image details by suppressing unintended changes than existing methods.

• Journal of Nutrition and Health
• /
• 제43권5호
• /
• pp.453-462
• /
• 2010

The effects of unripe apple mask pack made from unripe apple extracts on the skin status of middle-aged women for 4 weeks were investigated. The subjects were divided into three groups: placebo mask pack as the control (PM, n = 8), apple mask pack (AM, n = 16), and apple mask pack + apple intake (AAM, n = 12). The study included a survey questionnaire, and evaluations of nutrients intakes and of skin status. Skin status was measured by Aramo-TS. After 4 weeks, facial moisture, elasticity, evenness, and wrinkle level were improved in AM and AAM. The facial moisture level (AM: $+3.13\;{\pm}\;2.73$, AAM: $+2.25\;{\pm}\;2.93$) and elasticity level (AM: $+5.81\;{\pm}\;8.27$, AAM: $+5.50\;{\pm}\;6.13$) of AM (p < 0.001) and AAM (p < 0.05) were increased. The facial evenness level (AM: $-5.25\;{\pm}\;8.13$, AAM: $-9.17\;{\pm}\;4.26$) and wrinkle level (AM: $-10.88\;{\pm}\;11.74$, AAM: $-17.83\;{\pm}\;12.22$) were significantly decreased. AAM was more effective in improving skin elasticity, evenness, and wrinkle. The skin status of PM was not changed. While the macronutrient intakes of all the subjects were not different before and after the study, the daily intakes of some vitamins (vitamins $B_6$, C, and E and folic acid) during the study were significantly increased in AAM. These findings suggested that regular use of apple mask pack was effective in improving overall skin status. Furthermore, the consumption of apples with mask pack application was much better in improving skin status.

• 한국산업보건학회지
• /
• 제27권4호
• /
• pp.269-282
• /
• 2017

Objectives: Currently, masks against yellow dust and fine particulates are being certified with no consideration of facial dimensional variations among children and adults. The aims of this study were to develop masks against yellow dust and fine particulates for children in Korea and provide basic data to suggest new test methods for mask certification that consider the breathing capacity of children. Methods: A total of 730 study participants aged from six to 13 years old were recruited in the Seoul, Gyeonggi, and Incheon region. This study used a 3D scanning instrument to obtain 16 facial anthropometric data points. Literature reviews, a comparison of breathing capacity between adults and children, and analysis of children's pulmonary physiological data were conducted in order to suggest new test standards for certifying children's masks against yellow dust and fine particulates. In addition, types of children's masks, choice of wearing a mask or not, and reasons for not wearing masks were surveyed. Results: Based on a clustering analysis of participants' facial dimensions, facial shapes were classified into three groups: small, medium, and large. The sizes of children's masks were subtracted by using 3D sketch techniques(Large: $121.25mm{\times}89.46mm$, Medium: $111.92mm{\times}78.55mm$, Small: $102.13mm{\times}72.87mm$). In certifying children's mask, flow rates of $60{\ell}/m$ for the filtering efficiency test and $20{\ell}/m$ for the breathing resistance test were recommended, since children's pulmonary physiological capacity is about 60-70% of adults' pulmonary capacity. Conclusions: The results of this study suggest that three mask sizes for children would be sufficient and practical for providing protection against yellow dust and fine particulates. Revising current test methods for certifying respiratory protective devices for children is important, since children's pulmonary physiological capacity substantially differs from that of adults. Therefore, it is recommended that new test standards for certifying children's masks be promulgated in the near future.

• 대한안전경영과학회지
• /
• 제12권4호
• /
• pp.87-92
• /
• 2010

The mask-face interface design should consider the face shape to improve the half mask respirator's fit ratio. This study tried to design the mask-face interface using recent Korean face data. By using the data of 1536 men's 3D face scanning (Size Korea data), head clay mock-up was made and mask-face interface line was extracted from this head mock-up. Using this interface line, the half-mask prototype was made. According to the quantitative fitting test, the proposed mask was found to be well fitted (average fit-ratio > 100). The proposed method had two advantages. 1) The method could use massive head-related anthropometric data like Size Korea data. 2) The qualitative fit test (observation) could be conducted very quickly by fitting the prototype to the head mock-up. However, this method also had several limitations. 1) The head clay mock-up could be different according to the mock-up maker. 2) The average values of the head-related anthropometric data were used to make the head mock-up. Small and large size head mock-ups should be made and tested.

• Transactions on Control, Automation and Systems Engineering
• /
• 제3권4호
• /
• pp.283-288
• /
• 2001

In this paper, we proposed a method for extracting facial characteristics of human being in an image. Given a pair of gray level sample images taken with and without human being, the face of human being is segmented from the image. Noise in the input images is removed with the help of Gaussian filters. Edge maps are found of the two input images. The binary edge differential image is obtained from the difference of the two input edge maps. A mask for face detection is made from the process of erosion followed by dilation on the resulting binary edge differential image. This mask is used to extract the human being from the two input image sequences. Features of face are extracted from the segmented image. An effective recognition system using the discrete wave let transform (DWT) is used for recognition. For extracting the facial features, such as eyebrows, eyes, nose and mouth, edge detector is applied on the segmented face image. The area of eye and the center of face are found from horizontal and vertical components of the edge map of the segmented image. other facial features are obtained from edge information of the image. The characteristic vectors are extrated from DWT of the segmented face image. These characteristic vectors are normalized between +1 and -1, and are used as input vectors for the neural network. Simulation results show recognition rate of 100% on the learned system, and about 92% on the test images.

• 대한구순구개열학회지
• /
• 제4권2호
• /
• pp.9-18
• /
• 2001

구순구개열은 악안면 영역의 선천기형 중 발생빈도가 가장 높으며 출생시부터 성장완료 시기까지 일관된 치료계획에 의한 종합적 진료체계가 필수적이다. 구순구개열의 개선을 위한 구개부 및 입숭? 조기 봉합 후의 반흔조직의 형성은 치열궁 협착 뿐 아니라 상악골 열성장에 의한 중안모의 함몰을 야기하며 이러한 상악골의 열성장 경향은 연령이 증가할수록 점진적으로 증가하는 성향을 보인다. 따라서, 성장잠재력을 이용한 조기의 악정형치료는 구순구개열 환자에서 반드시 필요하며, 적절한 치료 결과를 얻어내기 위하여 환자의 전후방 골격 형태 뿐 아니라, 성장 방향에 대한 고려가 필요하다. 그러나 혼합치열기의 악정형치료는 좋은 결과를 얻었다 하더라도 높은 재발 성향 및 잔여성장에 의한 악골관계의 변화가 예상되므로 장기간의 보정이 필수적이며 성장에 대한 지속적 평가가 이루어져야 할 것이다.

• 한국미생물·생명공학회지
• /
• 제50권1호
• /
• pp.95-101
• /
• 2022

Due to the pandemic caused by COVID-19, the demand for face masks is soaring and has often caused a shortage. The aim of this study was to evaluate the effect of ultraviolet (UV) and drying treatments on microbial contaminants in facial masks. To conduct this study, standard procedures were designed to develop samples contaminated by the control bacteria Escherichia coli, Staphylococcus aureus, and Pseudomonas aeruginosa. The contamination level of the standard samples was approximately 6.30 × 10⁶ CFU/ml, and the UV light treatment was performed 1, 3, 5, and 7 times. To evaluate the effect of the UV and drying treatments, the masks were first treated with UV 1, 2, and 3 times, followed by the drying process. As a result, the mask contaminated with E. coli and P. aeruginosa showed a bacterial rate of approximately 99.9% after 1 UV irradiation, and in the case of the S. aureus-contaminated mask, it exhibited a bactericidal rate of approximately 99.9% after 7 UV irradiations. However, when the drying process was included after UV irradiation, all the samples contaminated with E. coli, S. aureus, and P. aeruginosa showed a bactericidal rate of 99.9% or more. The results of this study suggest that UV and drying treatments can effectively reduce the bacterial contaminants in facial masks. In addition, these results provide fundamental data and appropriate sterilization methods for reusing masks.