• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.03a
• /
• pp.157-160
• /
• 1998

본 논문에서는 동영상으로부터 에지 정보와 형판 변형을 통해 얼굴의 추출하고 그 특징을 기반으로 하는 추적 기법을 제안한다. 본 논문에서 제안하는 추적기법은 추출된 특징에 기반을 추적 지법으로 초기에 모자익 영상을 이용하여 얼굴 부분을 찾고 찾아진 얼굴 부분에 에지 연산자를 적용하여 에지를 추출한다. 에지 영상이 얻어지면 에지 영상에서 영역의 크기와 모양, 그리고 관계 검증을 통해 대략적인 눈 영역을 추출한다. 눈 영역이 찾아지면 이를 바탕으로 입 영역에 대한 후보 영역에 대하여 이진화를 수행하고 히스토그램 프로젝션을 통해 대략적인 입 영역을 추출한다. 추출된 눈 영역과 입 영역에 각각의 형판을 사용해 형판 변형을 하고 초기 매개변수를 추출한다. 추출된 매개변수는 다음 프레임에서 형판의 초기 값으로 사용된다. 그리고 나서 형판에 대하여 변형(deformation) 과정을 수행한다. 이 과정을 반복함으로써 추적 과정을 수행한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.574-576
• /
• 2003

현재 홍채 인식은 주로 수동형 영상 획득 시스템을 통한 홍채 획득이 주를 이루고 있다. 이는 장비가 고가인 점과 정확한 홍채 위치추적의 어려움 등의 문제로 인한 것이다. 본 연구에서는 24bit 칼라 영상에서 피부색 정보와 윤곽선 검출 정보를 이용한 실시간 자동 홍채 인식 시스템을 제안하였다. 제안한 방법에서는 HSI 칼라 좌표계상에서의 얼굴 피부색 인식 외에 조명으로 인한 잡음을 제거 하였고, 배경과 사용자의 보다 정확한 영역 분리를 위하여 영상을 이진화한 후 윤곽선 영역을 다시 한 번 제거 한 후 레이블링을 실행 하였다. 또한, 보다 정확한 눈 영역 추출을 위하여 일정 크기까지의 줌을 한 후 윤곽선 검출을 사용하였다. 이러한 방법들을 통하여, 주위 환경에 영향을 덜 받으면서 보다 정확한 눈 영역을 추출 할 수 있었다.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.12
• /
• pp.3675-3682
• /
• 1999

This paper proposes the method to extract face components without using the color information and the motion information in a gray image. A laplacian pyramid of the original image is built. Eye and nose candidates are extracted using only the gray information in a low resolution laplacian image and pairs are found that consist of two eye candidates and a nose one. At full resolution, horizontal and vortical edges are found in the regions of face components which are established using the candidates. Using those edge informations, face components are extracted. The experiments have been performed for images with various sizes and positions of face, and show very encouraging result.

• Journal of Korean Ophthalmic Optics Society
• /
• v.20 no.3
• /
• pp.277-284
• /
• 2015

Purpose: The study aimed to evaluate the effect of circle contact lens wearing on visual quality of the eyes after a refractive surgery. Methods: The objective visual quality was evaluated for 40 eyes who did not get refractive surgery and 30 eyes who got a refractive surgery after applying plano circle contact lenses on their eyes, respectively. Modulation transfer function (MTF), objective scatter index (OSI) and the focusing ratio on retina (Strehl ratio) were measured by using optical quality analysis system (OQAS) and the correlation between pupil size and objective visual quality was analyzed by measuring the pupil size. Results: When wearing circle contact lens on the eyes after refractive surgery, MTF and Strehl ratio were reduced and OSI was increased compared with the eyes without refractive surgery. The eyes after a refractive surgery showed more significant difference changes in MTF, OSI and Strehl ratio according to the pupil size compared with those without refractive surgery. Conclusions: The results showed that wearing of circle contact lens after the refractive surgery has deteriorated the objective visual quality based on the quality of focused image on the retina. Therefore, we suggest that sufficient understanding and consideration about the deterioration of visual quality is necessary in the case of circle contact lens wearing after refractive surgery.

• Journal of Korean Ophthalmic Optics Society
• /
• v.12 no.3
• /
• pp.35-38
• /
• 2007

This study surveyed on eye size, face types and glasses wears of normal 100 young peoples from 20 to 30 years old. The analysis of data resulted in as follows: 1. The vertical eye size were in case of below 2.4 mm (2%), 2.5-2.9 mm (40%), 3.0-3.5 mm (57%), and in case of high 3.6 mm (1%). 2. The horzontal eye size were in case of below 1.1 mm (52%), 1.2-1.4 mm (28%), 1.5-1.7 mm (18%), and in case of high 1.8 mm (2%). 3. As a whole, the face types were oval type (29%), round type (24%), square type (18%), and triangle type (2%). 4. As the male, the face types were oval type (32%), round type (10%), square type (26%), and triangle type (32%). 5. As the female, the face types were oval type (18%), round type (40%), square type (12%), and triangle type (30%). 6. As a whole, the FPD size were in case of larger than PD (49%), in case of suitable (25%), in case of smaller than PD (49%). 7. As the male, the FPD size were in case of larger than PD (28%), in case of suitable (36%), and in case of smaller than PD (32%). 8. As the female, the FPD size were in case of larger than PD (44%), in case of suitable (26%), and in case of smaller than PD (30%).

• Journal of the Korean earth science society
• /
• v.41 no.1
• /
• pp.1-18
• /
• 2020

This study presented the effects of the assumed mass-size relationship for snow on the simulated surface precipitation by using cloud microphysics parameterizations in Weather Research and Forecasting (WRF) model. The selected cloud microphysics parameterizations are WRF Double-Moment 6-class (WDM6) and WRF Single-Moment 6-class (WSM6) in the WRF model. We replaced the mass-size relationship for snow in WDM6 and WSM6 with Thompson's mass-size relationship retrieved from measurement data. The sensitivity of the modified WDM6 and WSM6 was tested for the idealized 2-dimensional squall line and winter precipitation system over the Korean peninsula, respectively. The modified WDM6 and WSM6 resulted in the increase of graupel/rain mixing ratios and the decrease of snow mixing ratio in the low atmosphere. The changes of hydrometeor mixing ratio and surface precipitation could be due to the collision-coalescence process between raindrops and snow and the graupel melting process.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1998.11a
• /
• pp.569-573
• /
• 1998

본 논문에서는 움직임 보상과 웨이브릿 변환 부호화 기법을 이용하여 비디오 신호의 데이터를 압축하고자 한다. 비디오 신호에 대한 인간 시각 특성을 분석하여 큰 이득을 주지 못하는 과정을 생략하고 인간의 눈에 민감하지 않은 정보를 생략함으로써 화질의 저하 없이 비디오 신호를 높은 비율로 압축한다. 제안한 방법은 움직임이 발생되는 정도와 크기에 매우 의존적으로, 움직임이 많은 비디오 신호에 더욱 적합하나 움직임이 작은 경우에도 저주파 대역들의 움직임 보상에 의한 이득이 크기 때문에 모든 영상에 대해서 적용 가능한 방법이다. 'football'과 'Popple'을 대상으로 한 컴퓨터 시뮬레이션결과, 프레임당 30:1 정도의 압축하에서 높은 재생화질을 얻을 수 있었다.

• The Journal of the Korean dental association
• /
• v.26 no.12 s.235
• /
• pp.1091-1095
• /
• 1988

안면의 미에 대한 연구는 최근에 이르기 까지 화가,조각가,철학가들의 영역이었다고 볼수 있다. 이에 대한 판정으느 주관적이었으며 관찰자의 눈에 의해 정의되었다. 그러나 임상적으로 안모형태를 변화시킬 수 있는 능력을 가지게 됨에 따라 보다 객관적인 미적 기준이 필요하게 되었다. 안면구조의 크기 형태와 내적관계에 대한 표준이 환자가 평균적으로 얼마나 벗어나고 어느정도로 임상적 중재를 해야 하는지 정하는데 필요했다. 정상적인 안모구조가 상실된 상황에서는 구조전 크기와 안모구조들의 정확한 위치에 대한 지식을 요구했다. 본 란에서는 심미적 안모를 양적으로 측정하는 일반기준과 교정학적 안모형태를 계측하는 방법을 소개하고자 한다.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.44 no.1
• /
• pp.56-63
• /
• 2007

In this paper, an eye tracking method is presented using a neural network (NN) and mean-shift algorithm that can accurately detect and track user's eyes under the cluttered background. In the proposed method, to deal with the rigid head motion, the facial region is first obtained using skin-color model and con-nected-component analysis. Thereafter the eye regions are localized using neural network (NN)-based tex-ture classifier that discriminates the facial region into eye class and non-eye class, which enables our method to accurately detect users' eyes even if they put on glasses. Once the eye region is localized, they are continuously and correctly tracking by mean-shift algorithm. To assess the validity of the proposed method, it is applied to the interface system using eye movement and is tested with a group of 25 users through playing a 'aligns games.' The results show that the system process more than 30 frames/sec on PC for the $320{\times}240$ size input image and supply a user-friendly and convenient access to a computer in real-time operation.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.2
• /
• pp.31-43
• /
• 2000

This paper presents a novel algorithm lot extraction of the eye and mouth fields (facial features) from 2D gray level face images. First of all, it has been found that Eigenfeatures, derived from the eigenvalues and the eigenvectors of the binary edge data set constructed from the eye and mouth fields are very good features to locate these fields. The Eigenfeatures, extracted from the positive and negative training samples for the facial features, ate used to train a MultiLayer Perceptron(MLP) whose output indicates the degree to which a particular image window contains the eye or the mouth within itself. Second, to ensure robustness, the ensemble network consisting of multiple MLPs is used instead of a single MLP. The output of the ensemble network becomes the average of the multiple locations of the field each found by the constituent MLPs. Finally, in order to reduce the computation time, we extracted the coarse search region lot eyes and mouth by using prior information on face images. The advantages of the proposed approach includes that only a small number of frontal faces are sufficient to train the nets and furthermore, lends themselves to good generalization to non-frontal poses and even to other people's faces. It was also experimentally verified that the proposed algorithm is robust against slight variations of facial size and pose due to the generalization characteristics of neural networks.