AAM 기반 얼굴 표정 인식을 위한 입술 특징점 검출 성능 향상 연구

한은정,강병준,박강령,Han, Eun-Jung,Kang, Byung-Jun,Park, Kang-Ryoung 한국정보처리학회 2009 정보처리학회논문지B Vol.16 No.4

AAM(Active Appearance Model) is an algorithm to extract face feature points with statistical models of shape and texture information based on PCA(Principal Component Analysis). This method is widely used for face recognition, face modeling and expression recognition. However, the detection performance of AAM algorithm is sensitive to initial value and the AAM method has the problem that detection error is increased when an input image is quite different from training data. Especially, the algorithm shows high accuracy in case of closed lips but the detection error is increased in case of opened lips and deformed lips according to the facial expression of user. To solve these problems, we propose the improved AAM algorithm using lip feature points which is extracted based on a new lip detection algorithm. In this paper, we select a searching region based on the face feature points which are detected by AAM algorithm. And lip corner points are extracted by using Canny edge detection and histogram projection method in the selected searching region. Then, lip region is accurately detected by combining color and edge information of lip in the searching region which is adjusted based on the position of the detected lip corners. Based on that, the accuracy and processing speed of lip detection are improved. Experimental results showed that the RMS(Root Mean Square) error of the proposed method was reduced as much as 4.21 pixels compared to that only using AAM algorithm. AAM(Active Appearance Model)은 PCA(Principal Component Analysis)를 기반으로 객체의 형태(shape)와 질감(texture) 정보에 대한 통계적 모델을 통해 얼굴의 특징점을 검출하는 알고리즘으로 얼굴인식, 얼굴 모델링, 표정인식과 같은 응용에 널리 사용되고 있다. 하지만, AAM알고리즘은 초기 값에 민감하고 입력영상이 학습 데이터 영상과의 차이가 클 경우에는 검출 에러가 증가되는 문제가 있다. 특히, 입을 다문 입력얼굴 영상의 경우에는 비교적 높은 검출 정확도를 나타내지만, 사용자의 표정에 따라 입을 벌리거나 입의 모양이 변형된 얼굴 입력 영상의 경우에는 입술에 대한 검출 오류가 매우 증가되는 문제점이 있다. 이러한 문제점을 해결하기 위해 본 논문에서는 입술 특징점 검출을 통해 정확한 입술 영역을 검출한 후에 이 정보를 이용하여 AAM을 수행함으로써 얼굴 특징점 검출 정확성을 향상시키는 방법을 제안한다. 본 논문에서는 AAM으로 검출한 얼굴 특징점 정보를 기반으로 초기 입술 탐색 영역을 설정하고, 탐색 영역 내에서 Canny 경계 검출 및 히스토그램 프로젝션 방법을 이용하여 입술의 양 끝점을 추출한 후, 입술의 양 끝점을 기반으로 재설정된 탐색영역 내에서 입술의 칼라 정보와 에지 정보를 함께 결합함으로써 입술 검출의 정확도 및 처리속도를 향상시켰다. 실험결과, AAM 알고리즘을 단독으로 사용할 때보다, 제안한 방법을 사용하였을 경우 입술 특징점 검출 RMS(Root Mean Square) 에러가 4.21픽셀만큼 감소하였다.

청피반성 혈관병증 치험 1례

한은정,김찬중,홍승욱,김윤범,Han, Eun-Jeung,Kim, Chan-Jung,Hong, Seung-Ug,Kim, Yoon-Bum 대한한방안이비인후피부과학회 2001 한방안이비인후피부과학회지 Vol.14 No.2

Livedo vasculopathy clinically shows purpuric papules and recurrent ulcers in the lower extremities, mainly on the ankles, leaving characteristic scars called atrophie blanche after the healing of the ulcers. In Korean literture, about 10 cases of this disease have been reported but nothing in korean oriental medicine. So, we report a case of livedo vasculopathy in a 46 year-old female patient treated by oriental medicine therapy.

치아미백에서 치위생사의 역할

한은정,Han, Eun-Jeong 대한심미치과학회 2005 Journal of the Korean Academy of Esthetic Dentistr Vol.14 No.1

알로에(蘆회), 녹차(綠茶), 꿀(蜂蜜)의 미백효과에 관한 연구

한은정,이길영,김혜정,김윤범,Han, Eun-jeong,Lee, Gil-young,Kim, Hae-jeong,Kim, Yoon-bum 대한한방안이비인후피부과학회 2003 한방안이비인후피부과학회지 Vol.16 No.3

Objectives : This study was performed to investigate the depigmentation effects of Aloe, Camellia sinensis and Mel. Methods : Inhibition of tyrosinase activity, melanin production & melanoma cell viability in cultured B16 melanoma cells, UV screen and cytoprotective effects on PC12 cells injured by hydrogen peroxide were measured. Results : Aloe has some inhibitory effects on tyrosinase activity, on the other hand Camellia sinensis and Mel do not have. They did not show any inhibitory effects on melanin production in melanoma cells and cytoprotective effects on PC12 cells injured by hydrogen peroxide. Aloe and Camellia sinensis have some inhibitory effects on UV screen. Conclusions : This study shows that Aloe and Camellia sinensis which were generally used for external application have some depigmentation effects. Following this, We should use them for whitening agents and the depigmentation effects of the other natural subjects which were generally used for external application should be examined.

노인장기요양보험 이용지원 상담 대상자 선정모형 개발

한은정,김동건,Han, Eun-Jeong,Kim, Dong-Geon 한국통계학회 2015 응용통계연구 Vol.28 No.6

In the long-term care insurance (LTCI) system, National Health Insurance Service (NHIS) provide counseling services for beneficiaries and their family caregivers, which help them use LTC services appropriately. The purpose of this study was to develop a Target Selection Model for the Counseling Services based on needs of beneficiaries and their family caregivers. To develope models, we used data set of total 2,000 beneficiaries and family caregivers who have used the long-term care services in their home in March 2013 and completed questionnaires. The Target Selection Model was established through various data-mining models such as logistic regression, gradient boosting, Lasso, decision-tree model, Ensemble, and Neural network. Lasso model was selected as the final model because of the stability, high performance and availability. Our results might improve the satisfaction and the efficiency for the NHIS counseling services. 우리나라 노인장기요양보험에서는 수급자와 그 가족부양자가 수급자의 심신기능 상태와 욕구에 따라 불이익이나 불편함이 없이 비용-효과적으로 장기요양 급여를 이용할 수 있도록 지원하고자 이용지원 상담을 제공하고 있다. 본 연구는 재가급여 이용자의 이용지원 정기상담 대상자 선정시 상담 대상자의 욕구를 반영하지 않아 이용지원 상담의 만족도와 효율성이 낮은 문제를 통계학적 모형을 활용하여 해결하고자 수행되었다. 모형 개발을 위해 2013년 3월 장기요양 재가급여를 이용한 수급자와 가족부양자를 대상으로 이용지원 상담에 대한 욕구와 관련 변수를 조사하였으며, 2,000명이 조사를 완료하였다. 조사 자료를 바탕으로 이용지원 상담 대상자 선정모형을 다양한 데이터마이닝 기법(로지스틱 회귀모형, 의사결정 나무모형, Lasso 모형, 자동 신경망모형, 그래디언트 부스팅, 앙상블 모형)을 통해 개발하였고, 이중 가장 안정적이고 현장 적용이 쉽고 성능이 좋은 Lasso 모형 결과를 최종모형으로 선정하였다. 본 연구가 이용지원 상담의 만족도를 높이고 업무를 효율화 하는데 기여할 것으로 기대된다.

Random Forests 기법을 이용한 백내장 예측모형 - 일개 대학병원 건강검진 수검자료에서 -

한은정,송기준,김동건,Han, Eun-Jeong,Song, Ki-Jun,Kim, Dong-Geon 한국통계학회 2009 응용통계연구 Vol.22 No.4

백내장 질환은 노령인구가 증가하고 있는 시점에서 사회, 경제적으로 심각한 문제로 부각되고 있는 질병으로 조기 진단이 이루어진다면 발병률을 크게 줄일 수 있는 질병이다. 본 연구에서는 백내장을 조기 진단하기 위한 예측 모형을 구축하고자 1994년부터 2001년까지 연세대학병원에서 2회 이상 건강검진을 받고 의사진단을 통해 백내장 여부를 확인할 수 있는 30세 이상 남 녀 3,237명에 대한 건강검진 수검 자료를 활용하여 백내장 발생 위험 예측모형을 개발하였다. 모형개발에는 데이터마이닝 기법인 Random Forests를 사용하였고, 기존의 로지스틱 회귀분석, 판별분석, 의사결정나무 모형(Decision tree), 나이브베이즈(Naive Bayes), 앙상블 모형인 배깅(Bagging)과 아킹(Arcing)을 이용하여 그 성능을 비교 분석하였다. Random Forests를 통해 개발한 백내장 발생 예측모형은 정확도가 67.16%, 민감도가 72.28%였고, 주요 영향요인은 연령, 혈당, 백혈구수치(WBC), 혈소판수치(platelet), 중성지질(triglyceride), BMI였다. 이 결과는 의사의 안과검진 정보 없이 건강검진 수검 자료만으로 백내장 질환 유 무에 관한 정보를 70% 정도 예측할 수 있음을 보여주는 것으로, 백내장의 조기 진단에 많은 기여를 할 것으로 판단된다. Cataract is the main cause of blindness and visual impairment, especially, age-related cataract accounts for about half of the 32 million cases of blindness worldwide. As the life expectancy and the expansion of the elderly population are increasing, the cases of cataract increase as well, which causes a serious economic and social problem throughout the country. However, the incidence of cataract can be reduced dramatically through early diagnosis and prevention. In this study, we developed a prediction model of cataracts for early diagnosis using hospital data of 3,237 subjects who received the screening test first and then later visited medical center for cataract check-ups cataract between 1994 and 2005. To develop the prediction model, we used random forests and compared the predictive performance of this model with other common discriminant models such as logistic regression, discriminant model, decision tree, naive Bayes, and two popular ensemble model, bagging and arcing. The accuracy of random forests was 67.16%, sensitivity was 72.28%, and main factors included in this model were age, diabetes, WBC, platelet, triglyceride, BMI and so on. The results showed that it could predict about 70% of cataract existence by screening test without any information from direct eye examination by ophthalmologist. We expect that our model may contribute to diagnose cataract and help preventing cataract in early stages.

긴삼의 db/db 마우스에서 항당뇨 효과

한은정,박금주,최윤숙,한기철,박종석,이경희,고성권,정성현,Han, Eun-Jung,Park, Keum-Ju,Choi, Yun-Sook,Han, Ki-Chul,Park, Jong-Suk,Lee, Kyung-Hee,Ko, Sung-Kwon,Chung, Sung-Hyun 대한약학회 2006 약학회지 Vol.50 No.5

Type 2 diabetes mellitus is a chronic and hard to control disease. In order to develop the therapeutic agent for type 2 DM, many researchers investigated natural products using an in vitro and in vivo assay. In this study, we tried to explore the anti-diabetic activity and mechanisms of ginsam, which is a vinegar-processed ginseng radix. The db/db mice were randomly divided into four groups. The diabetes control (DC) group was orally administrated with distilled water, ginseng radix (GR) or ginsam (GS) was administrated orally at a dose of 150 mg/kg, and the positive control group was orally injected with metformin (MET) at a dose of 300 mg/kg for 5 weeks in db/db mice and measured body weight and blood glucose level every week. All treatment groups decreased the plasma glucose levels compared with diabetic control and GS group significantly lowered the insulin resistance index. GS group also reduced the plasma lipid levels mainly due to reduce the lipogenesis and increase the lipolysis in the fat tissue. In addition, GS group increased the GLUT4 mRNA expression levels in the fat and muscle tissues by 10 fold probably due to increase a $PPAR_{-\gamma}$ mRNA expression in fat tissue. Taken together, GS showed the anti-hyperglycemic and anti-hyperlipidemic activities and those activities may ascribe to over-expression of GLUT4 mRNA level and decrease the lipogenesis in fat tissue.

鼻症狀을 동반한 Maxillary Sinus Retention Cyst 1예

한은정,김윤범,Han, Eun-jeong,Kim, Yoon-bum 대한한방안이비인후피부과학회 2003 한방안이비인후피부과학회지 Vol.16 No.1

Maxillary sinus retention cysts are probably common findings on PNS series. Many patients remain asymptomatic in the presence of these lesions, but a variety of local and systemic symptoms have been associated with them. A patient presented with symptoms of chronic rhinosinusitis had a finding of both maxillary sinus retention cysts on PNS series. He has taken the Gal-gen-tang and treated by acupunture & infra-red. After about 4 months' treatment, his symptoms resolved and PNS series were normal. In conclusion, Maxillary sinus retention cysts may become clinically important when they occur in the setting of symptoms compatible with chronic rhinosinusitis, or when the diagnosis is in doubt.

전작물 재배에 의한 마늘 흑색썩음균핵병 억제효과

한은정,최재필,김용기,홍성준,박종호,심창기,김민정,김석철,윤석한,Han, Eun-Jung,Choi, Jae-Pil,Kim, Yong-Ki,Hong, Sung-Jun,Park, Jong-Ho,Shim, Chang-Ki,Kim, Min-Jeong,Kim, Seok-Cheol,Yoon, Seok-Han 한국유기농업학회 2015 韓國有機農業學會誌 Vol.23 No.1

본 연구는 전작물 재배에 따른 마늘 흑색썩음균핵병 피해 경감 효과를 평가하기 위해 2 년간 수행되었다. 콩 참깨 호박 네마장황, 대파를 전작물로 재배하였으며, 대조구로 휴경처리구와 병원균 무접종구를 두었다. 전적물 재배 후 마을을 파종하고 재배하면서, 흑색썩음균핵병 이병주율과 수량을 조사하였다. 2013년 흑색썩음균핵병은 병원균 무접종구, 콩, 네마장황처리구에서 가장 낮았으며, 호박처리구에서 가장 높았다. 2014년에는 네마장황, 콩, 대파처리구에서 흑색썩음균핵병 이병주율이 낮았으며, 호박처리구에서 가장 높았다. 처리구 별 마늘의 수량은 2013년에는 콩, 네마장황, 대파 처리구에서 가장 높았고 호박처리구에서 가장 적었으며, 2014년에는 콩, 네마장황, 대파처리구에서 높았고, 호박처리구에서 가장 적어 이병주율과 유사한 경향을 보였다. 이상의 결과를 토대로 콩-마늘과 네마장황-마늘재배체계가 흑색썩음균핵병을 억제하기 위한 윤작체계가 될 수 있을 것으로 생각되었다. This study was conducted to evaluate the effect of crop rotation cultivation on the suppression of garlic white rot caused by Sclerotium cepivorum in the mini plot ($2^*1^*$ 0.5 m). Six crops, soybean, sesame, mung bean, squash, crotalaria and spring onion, were previously transplanted in the mini-plots infested with S. cepivorum before garlics were planted. After cultivation of the previous crops, garlic was sown in the mini-plot. Non-cultivation plots and non-infested plots with white rot pathogen were used as control. The effect of crop rotation cultivation on the suppression of garlic white rot was evaluated by investigating comparatively the disease incidence (the percentage of infected plants) and yields. As a results, infection rate of garlic white rot was recorded lower in the non-infested plot, crotalaria and soybean cultivation than in the plot of the other crop cultivation. Especially when squash was previously cultivated and garlics were planted in 2013, infection rate of garlic white was recorded the highest score. In 2014, the infection rate of garlic white were low in the garlic on soybean, crotalaria and spring onion treatment whereas it was high in squash treatment, as well. In 2013, garlic yield was the highest in no inoculation plot, followed by crotalaria, soybean, no crop cultivation, sesami, mungbean, squash cultivation plot. In 2014, the yield in the plot of crotalaria and soybean was much higher than that in no inoculation plot. Based on above-described results, it is considered that soybean-garlic and crotalaria-garlic cultivation system can be good crop rotation systems to control garlic white rot.

PubMed 검색을 통해 살펴본 Aromatherapy의 임상연구동향에 관한 고찰

한은정,김윤범,Han, Eun-Jeung,Kim, Yoon-Bum 대한한방안이비인후피부과학회 2002 한방안이비인후피부과학회지 Vol.15 No.1

Aromatherapy is becoming increasingly popular, however there are few clear indications for its use. To systematically review the clinical studies on aromatherapy, computcrized literature searches were performed from PubMed. Fourteen trials were located. The conclusions were as follows: 1. All the trials were published after 1993. 2. To analyze the publication journals, five of fourteen trials were on nursing, three of them on neuropsychiatry and one on dermatology. 3. Most widely used indications of aromatherapy were on general physical, psychological and physiological field. 4. On the procedure of aromatherapy, eight of them were by massage, three by inhalation, one by massage & inhalation and one by bath additive. 5. To analyze the publication type, all fourteen trials were clinical trials. Twelve of them were randomized controlled trials. Two of them were using single subject design. 6. To observe the sample size, five of them were under 10, four from 10 to 50, two from 50 to 100 and three above 100. 7. Eleven of fourteen trials were using inferential statistics. Eight of the eleven trials were significantly effective, but three had no statistical significance.