• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.583-585
• /
• 2012

본 논문은 웹카메라를 이용하여 얼굴이 바라보고 있는 방향을 인식하는 시스템을 제안한다. 얼굴 검출 방법으로 Haar-like Face Detect를 이용하여 얼굴을 검출하고 전체 이미지에서 검출된 얼굴 영역만을 관심영역으로 설정하여 Haar-like Eye Detect를 이용하여 눈 영역을 검출하였다. 검출된 눈 위치에 대한 평균값으로 얼굴이 왼쪽 방향을 보고 있는지 오른쪽 방향을 보고 있는지를 판단하였다. 제안된 방법의 실험 결과, 얼굴 및 눈 영역을 비교적 정확하게 검출하였으며 계산된 눈 위치를 이용하여 얼굴 방향 인식에 대해서 우수한 성능을 보였다.

• Current Optics and Photonics
• /
• v.6 no.2
• /
• pp.196-201
• /
• 2022

Collagen fibers tens of nanometers in size, which constitute most of the corneal volume of the human eye, are layered in a uniform direction, and adjacent fiber layers are arranged at an angle of 90° to each other. According to the results of this study, the transmittance at 45° of interlayer rotation angle is highest, and higher than that of the 90° body structure. The transmittance is examined, concerning the polarization state of the incident light; circularly polarized light case shows higher transmittance than linearly polarized. Through this, a simulation to confirm the deformed structure of collagen fibers, which show higher transmittance than the anatomical structure of the cornea, is attempted.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2021.06a
• /
• pp.180-182
• /
• 2021

본 논문은 눈 랜드마크 위치 검출과 시선 방향 벡터 추정이 하나의 딥러닝 네트워크로 통합된 시선 추정 네트워크를 제안한다. 제안하는 네트워크는 Stacked Hourglass Network[1]를 백본(Backbone) 구조로 이용하며, 크게 랜드마크 검출기, 특징 맵 추출기, 시선 방향 추정기라는 세 개의 부분으로 구성되어 있다. 랜드마크 검출기에서는 눈 랜드마크 50개 포인트의 좌표를 추정하며, 특징 맵 추출기에서는 시선 방향 추정을 위한 눈 이미지의 특징 맵을 생성한다. 그리고 시선 방향 추정기에서는 각 출력 결과를 조합하고 이를 통해 최종 시선 방향 벡터를 추정한다. 제안하는 네트워크는 UnityEyes[2] 데이터셋을 통해 생성된 가상의 합성 눈 이미지와 랜드마크 좌표 데이터를 이용하여 학습하였으며, 성능 평가는 실제 사람의 눈 이미지로 구성된 MPIIGaze[3] 데이터 셋을 이용하였다. 실험을 통해 시선 추정 오차는 0.0396 MSE(Mean Square Error)의 성능을 보였으며, 네트워크의 추정 속도는 42 FPS(Frame Per Second)를 나타내었다.

• Journal of Internet Computing and Services
• /
• v.14 no.6
• /
• pp.117-124
• /
• 2013

In this paper, we present a new method which efficiently estimates a face direction from a sequences of input video images in real time fashion. For this work, the proposed method performs detecting the facial region and major facial features such as both eyes, nose and mouth by using the Haar-like feature, which is relatively not sensitive against light variation, from the detected facial area. Then, it becomes able to track the feature points from every frame using optical flow in real time fashion, and determine the direction of the face based on the feature points tracked. Further, in order to prevent the erroneously recognizing the false positions of the facial features when if the coordinates of the features are lost during the tracking by using optical flow, the proposed method determines the validity of locations of the facial features using the template matching of detected facial features in real time. Depending on the correlation rate of re-considering the detection of the features by the template matching, the face direction estimation process is divided into detecting the facial features again or tracking features while determining the direction of the face. The template matching initially saves the location information of 4 facial features such as the left and right eye, the end of nose and mouse in facial feature detection phase and reevaluated these information when the similarity measure between the stored information and the traced facial information by optical flow is exceed a certain level of threshold by detecting the new facial features from the input image. The proposed approach automatically combines the phase of detecting facial features and the phase of tracking features reciprocally and enables to estimate face pose stably in a real-time fashion. From the experiment, we can prove that the proposed method efficiently estimates face direction.

• Korean Institute of Interior Design Journal
• /
• v.26 no.5
• /
• pp.83-92
• /
• 2017

This research is about the characteristics of the saccades that occurring in the process of observing the spatial. The size of the saccades is the visual activity for acquiring the information, and the search trace that left by the spatial visitor. Reviewed over the saccade through the analysis of observation process that appeared in the eye-tracking experiment of sports store from the department store, the saccade is defined as the sight movement in the process of finding a component to view in the spatial. If some spatial component induces saccade, and able to know which specific component acquired through the saccade, then the designer will have a clue which able to provide the spatial desired by the consumer. The results of analyzing the process in the deriving the spatial component and saccade characteristic from industrial space, can be summarised as follows. The first, the average fixed number of [IN].[OUT] zone appeared similarly, and it leads to knowing that the time used for attention with one frequency is about 0.3 seconds. Second, there were more saccades toward [$IN{\rightarrow}OUT$] (13 persons, 76.5%) than [$OUT{\rightarrow}IN$] (4 persons, 23.5%). The Los Angeles area has lots of interesting things to watch, so it able to thinks to begins with small saccade and then occur to bigger saccade activity to find new interests shortly after [OUT] saccade. Third, according to time range changes, keep eyes on the characteristics of saccade, [IN].[OUT] frequency has slightly decreased the changes of viewing time, but there was no significant change in an average number of observations. This means that the frequency and the number of observations are decreased together. Therefore, it can be seen that the amount of information to be acquired (frequency) is decreased (count) as the observation time elapses.

• Archives of design research
• /
• v.11 no.3
• /
• pp.187-196
• /
• 1998

This study is intended to suggest the method of text arrangement in order to enhance visual perception which would help darify its communication. One hundred subjects without restriction of gender and profession participated in each experiment :their reading time was measured by the 0.01 second. The Analysis of Variance(two-way ANOVA without interaction) was performed for each experiment and the p value was 0.0001 which implies that there was a strong consistency among test results. Based on the first results, it is found that there is a consistent relationship between type size and text line length, and the following discovery was made ; the most effective ratio of type size to line length is approximately 1 :8. Judging from the Second and Third results, it seems that the vertical text arrangement is most efficient for reading regardless of text line length. So to make same rreading direction is more important than to narrow down the eye moving distance between column and column for efficient visual flow. This research supports the view that considering efficient eye movement on text, it is important to understand the mentioned variables that affect visual interpretation.

• Journal of Korean Ophthalmic Optics Society
• /
• v.6 no.2
• /
• pp.47-52
• /
• 2001

We investigate the variation in the shape of confusion circle as the real amplitude distribution of incident beam in pupil area is shifted and the effect of the shift of incident beam on the astigmatism for eye. The shifted magnitudes of Incident beam are 0.0, 0.25, 0.5 and 0.75 and the shifted direction of incident beam ${\pi}/2$. We also consider the optical system having the astigmatism which are $0.0{\lambda}$, $0.25{\lambda}$, $0.5{\lambda}$ and $0.75{\lambda}$. As the shifted magnitude of the real amplitude distribution of incident beam increases, the shape of confusion circle on the image surface transfers from the rotational symmetry to the asymmetry like ellipse and the length of the major axis for confusion circle on the Image surface increases. We know from results in this paper that the shift of the incident beam in pupil area compensates the effect of astigmatism and the real amplitude distribution of incident beam in pupil area is shifted to minimize the effect of astigmatism for eye.

• Journal of Haehwa Medicine
• /
• v.20 no.1
• /
• pp.25-37
• /
• 2011

It fuses into culture of the East and the West in use of color ; colors of five direction, which had been from the Five Phase Theory in East Asia. Formative ability of color fields in a formative early Oriental Medicine history as well. Color is a very important factor and a necessary step of visible diagnosis. As a human's body is maintained and alive under the control of spirit and spirit is stored by five viscera, so spirit expresses the change of five viscera and is reflected by color. Visible diagnosis consists of spirit, appearance, Qi and color and movement. The purpose of each visible diagnosis is that we would know states of essence, Qi and spirit in patient's body. To ancient Asia people 'to see' was a kind of insight to the object as a whole. Similarily the activity of seeing the human body was the integral part of making diagnosis of a patient. This was the cause that there was suggested the discussion of the Five color theory in Oriental Medicine as a counterpart to that of the 'Goethe for Zur Farbenlehrer'. The inspection of Oriental Medicine was not a simple gazing of the eye as a sense organ, but the total insight to the internal state of the patient. For that reason, the eye－perception in early Chinese medicine was the reading the signs of the internal body which had have not the visual form but the flux of the internal life.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.5
• /
• pp.187-195
• /
• 2018

In this study, using real-time gaze tracking and head tracking method, we intended to quantitatively evaluate the deviation between the patient's head and gaze direction while minimizing inspection errors due to apraxia of conventional paper-based examination respectively. As a result, we developed a software that can quantitatively measure gaze and head movement information, and computerized the line bisection and star cancelation test, which are generally used as conventional paper test. In addition, for the rehabilitation training, contents corresponding to the visual technology of Warren's visual hierarchical model lower level are implemented and can be performed repetitively and independently. This allows the patient to actively participate in rehabilitation and quantitatively compare the degree of improvement.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.3
• /
• pp.210-216
• /
• 2016

This paper presents a novel collision avoidance technique for mobile robots based on omni-directional vision simultaneous localization and mapping (SLAM). This method estimates the avoidance path and speed of a robot from the location of an obstacle, which can be detected using the Lucas-Kanade Optical Flow in images obtained through fish-eye cameras mounted on the robots. The conventional methods suggest avoidance paths by constructing an arbitrary force field around the obstacle found in the complete map obtained through the SLAM. Robots can also avoid obstacles by using the speed command based on the robot modeling and curved movement path of the robot. The recent research has been improved by optimizing the algorithm for the actual robot. However, research related to a robot using omni-directional vision SLAM to acquire around information at once has been comparatively less studied. The robot with the proposed algorithm avoids obstacles according to the estimated avoidance path based on the map obtained through an omni-directional vision SLAM using a fisheye image, and returns to the original path. In particular, it avoids the obstacles with various speed and direction using acceleration components based on motion information obtained by analyzing around the obstacles. The experimental results confirm the reliability of an avoidance algorithm through comparison between position obtained by the proposed algorithm and the real position collected while avoiding the obstacles.