• 정보처리학회논문지B
• /
• 제8B권2호
• /
• pp.183-188
• /
• 2001

이 논문은 눈의 형태에 대한 정보를 이용하여 눈동자를 효과적으로 찾는 방법과 얼굴 특성을 반영하는 평가함수를 이용하여 눈동자, 입의 위치와 같은 얼굴 특징들을 인식하는 방법을 제안하였다. 색 정보를 이용하여 입술과 얼굴 영역을 추출하고 눈동자와 흰자위간의 명도 차를 이용하는 함수를 사용하여 눈동자를 인식하였다. 마지막으로 얼굴 특성을 반영하느 평가함수를 정의하고 이를 이용하여 최종적인 얼굴과 눈, 입을 인식하였다. 제안한 방법을 사용하여 여러 영상들에 대해 실험하여 좋은 결과를 얻었다.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2000년도 봄 학술발표논문집 Vol.27 No.1 (B)
• /
• pp.574-576
• /
• 2000

디지털 영상 처리에서 사람 얼굴 인식은 여러 응용 분야에 요구되어 활발한 연구가 진행되어 왔다. 본 논문에서는 얼굴이 있는 칼라 영상에서 얼굴 영역을 자동으로 추출하기 위한 알고리즘을 기술한다. 영상에 있는 얼굴 및 복잡한 배경과 다른 구성 성분들을 분류시켜 표현할 수 있도록 하기 위해 다변수 히스토그램 기법을 이용하여 얼굴 후보 영역과 머리 후보영역을 추출하고, 머리와 얼굴 후보 영역들간의 상관관계를 고려하여 최종 얼굴 영역을 검출한다. 본 논문에서 제안된 방법의 성능을 검증하기 위하여 인터넷상에 있는 128개의 영상을 입력 데이터로 실험한 결과, 인식 시간이 빠르고 영상의 배경에도 강건한(robust) 효율적인 방법임을 알 수 있었다.

• Investigative Magnetic Resonance Imaging
• /
• 제19권3호
• /
• pp.162-167
• /
• 2015

Purpose: To evaluate the usefulness of quantitative analysis of the facial nerve using contrast-enhanced three-dimensional (CE 3D) fluid-attenuated inversion recovery-volume isotopic turbo spin echo acquisition (FLAIR-VISTA) for the diagnosis of Bell's palsy in pediatric patients. Materials and Methods: Twelve patients (24 nerves) with unilateral acute facial nerve palsy underwent MRI from March 2014 through March 2015. The unaffected sides were included as a control group. First, for quantitative analysis, the signal intensity (SI) and relative SI (RSI) for canalicular, labyrinthine, geniculate ganglion, tympanic, and mastoid segments of the facial nerve on CE 3D FLAIR images were measured using regions of interest (ROI). Second, CE 3D FLAIR and CE T1-SE images were analyzed to compare their diagnostic performance by visual assessment (VA). The sensitivity, specificity, and accuracy of RSI measurement and VA were compared. Results: The absolute SI of canalicular and mastoid segments and the sum of the five mean SI (total SI) were higher in the palsy group than in the control group, but with no significant differences. The RSI of the canalicular segment and the total SI were significantly correlated with the symptomatic side (P = 0.028 and 0.015). In 11/12 (91.6%) patients, the RSI of total SI resulted in accurate detection of the affected side. The sensitivity, specificity, and accuracy for detecting Bell's palsy were higher with RSI measurement than with VA of CE 3D FLAIR images, while those with VA of CE T1-SE images were higher than those with VA of CE 3D FLAIR images. Conclusion: Quantitative analysis of the facial nerve using CE 3D FLAIR imaging can be useful for increasing the diagnostic performance in children with Bell's palsy when difficult to diagnose using VA alone. With regard to VA, the diagnostic performance of CE T1-SE imaging is superior to that of CE 3D FLAIR imaging in children. Further studies including larger populations are necessary.

• 디지털산업정보학회논문지
• /
• 제5권2호
• /
• pp.167-176
• /
• 2009

In this paper, a system that detects driver's drowsiness has been implemented based on the automatic extraction and the tracking of pupils. The research also focuses on the compensation of illumination and reduction of background noises that naturally exist in the driving condition. The system, that is based on the principle of Haar-like feature, automatically collects data from areas of driver's face and eyes among the complex background. Then, it makes decision of driver's drowsiness by using recognition of characteristics of pupils area, detection of pupils, and their movements. The implemented system has been evaluated and verified the practical uses for the prevention of driver's drowsiness.

• Journal of Multimedia Information System
• /
• 제2권1호
• /
• pp.171-178
• /
• 2015

This article describes 3 dimensional (3D) face recognition system using histogram of oriented gradients (HOG) based on face curvature. The surface curvatures in the face contain the most important personal feature information. In this paper, 3D face images are recognized by the face components: cheek, eyes, mouth, and nose. For the proposed approach, the first step uses the face curvatures which present the facial features for 3D face images, after normalization using the singular value decomposition (SVD). Fisherface method is then applied to each component curvature face. The reason for adapting the Fisherface method maintains the surface attribute for the face curvature, even though it can generate reduced image dimension. And histogram of oriented gradients (HOG) descriptor is one of the state-of-art methods which have been shown to significantly outperform the existing feature set for several objects detection and recognition. In the last step, the linear discriminant analysis is explained for each component. The experimental results showed that the proposed approach leads to higher detection accuracy rate than other methods.

• 제어로봇시스템학회논문지
• /
• 제15권4호
• /
• pp.379-384
• /
• 2009

A human gaze detection and tracing method is importantly required for HMI(Human-Machine-Interface) like a Human-Serving robot. This paper proposed a novel three-dimension (3D) human gaze estimation method by using a face recognition, an orientation estimation and SVMs (Support Vector Machines). 2,400 images with the pan orientation range of $-90^{\circ}{\sim}90^{\circ}$ and tilt range of $-40^{\circ}{\sim}70^{\circ}$ with intervals unit of $10^{\circ}$ were used. A stereo camera was used to obtain the global coordinate of the center point between eyes and Gabor filter banks of horizontal and vertical orientation with 4 scales were used to extract the facial features. The experiment result shows that the error rate of proposed method is much improved than Liddell's.

• 조명전기설비학회논문지
• /
• 제28권4호
• /
• pp.70-79
• /
• 2014

In this paper, I propose a method for detection of look-alike offenders by using hidden face information. For extraction of moving objects, PRA matching is used to extract moving components, and brightness changes can be dealt with by an adaptive threshold adjusting in the proposed method. Moving objects extracted in the territory of the face region is extracted using the complexion, facial area, eyes, nose, mouth. The extracted information detected by the presence of these characteristics were likely to help judge a person. Results of the extracted face makes the recognition rate of possible murderers 90% so the usefulness of the proposed method was confirmed.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1996년도 춘계학술대회
• /
• pp.56-59
• /
• 1996

The purpose of this study is to identify the cause of driver's drowsiness and to get information about driver's drowsiness from facial image using computer vision. We measured the driver's movements of a head and shoulders in the highway arid street. We also measured the eye blink duration and yawning duration of normal and drowsy drivers. from the results, we confirmed that the measurement of eye blink and yawning might be a way of drowsy detection.

• International Journal of Computer Science & Network Security
• /
• 제22권7호
• /
• pp.165-170
• /
• 2022

Lying is nuisance to all, and all liars knows it is nuisance but still keep on lying. Sometime people are in confusion how to escape from or how to detect the liar when they lie. In this research we are aiming to establish a dynamic platform to identify liar by using video analysis especially by calculating the ratio of changes in their appearance when they lie. The platform will be developed using a machine learning algorithm along with the dynamic classifier to classify the liar. For the experimental analysis the dataset to be processed in two dimensions (people lying and people tell truth). Both parameter of facial appearance will be stored for future identification. Similarly, there will be standard parameter to be built for true speaker and liar. We hope this standard parameter will be able to diagnosed a liar without a pre-captured data.

• 한국멀티미디어학회논문지
• /
• 제4권5호
• /
• pp.396-405
• /
• 2001

일반적으로 피부색과 유사한 칼라가 포함된 복잡한 배경으로부터 미리 정의된 단일 칼라 임계값을 이용해 얼굴영역을 정확하게 추출하는 것은 매우 어렵다. 이에 본 연구에서는 초기 피부색 범위를 점진적으로 확장시키면서 얼굴 영역을 추출하는 방법을 제안한다. 초기 피부색 범위는 인터넷에서의 다양한 영상으로부터 얼굴 영역의 피부색 분포를 분석하여 매우 밀집되어 나타나는 범위로 선정한다. 피부색 범위를 확장함에 따라 영상 공간에서의 얼굴후보 영역도 확장되어 나타나는데, 확장 단계별로 확장된 얼굴 후보 영역이 실제 얼굴임을 검증하기 위해 얼굴의 모양정보와 얼굴 기관의 위치적 정보를 사용한다. 모양 제한자로는 타원의 특성과 외접 사각형(bounding box)의 가로와 세로의 비율을 사용한다. 얼굴 기관으로는 눈과 입만을 사용했는데, 얼굴 후보 영역 내부의 가로 방향성을 가지는 경계가 주로 눈과 입의 영역임에 착안한 것으로서 비교적 수월하게 추출할 수 있기 때문이다. 실험을 통해, 제안한 방법이 하이라이트(highlight)에 의해 얼굴 영역의 일부가 왜곡된 경우와 얼굴 영역이 피부색과 유사한 배경에 인접해 있는 경우에 대해서도 강인하게 얼굴 영역을 추출할 수 있음을 확인할 수 있었다.