• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.06a
• /
• pp.356-358
• /
• 2016

본 논문에서는 동공 검출 기반 눈 검출 및 추적 시스템을 구현하기 위한 방법의 최적 조건을 제안한다. 이러한 시스템을 구현하려면, 구현하기 위해 사용하는 동공 명암 차 현상이 어떠한 환경과 시스템을 갖춰야 하는지에 대한 정보가 필요하다. 본 논문에서는 어떠한 조건에서 동공 명암 차 현상이 확실하게 검출이 되는지를 실험하였다. 실험결과 본 논문의 실험 환경내에서 카메라 렌즈와 눈 사이의 거리가 60cm에서 80cm사이일 때, 조명의 세기가 강할수록 동공 명암 차 현상이 강인하게 검출되는 것을 확인하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2013.11a
• /
• pp.112-113
• /
• 2013

서브 홀로그램 디스플레이는 디지털 홀로그래피 디스플레이의 제한된 시역을 관찰자의 동공 크기로 맞게 구현하여 사용자가 더 넓은 범위에서 더 큰 영상을 느끼도록 만드는 홀로그래픽 디스플레이이다. 본 논문에서는 서브홀로그램 방식에서 시야 창 문제를 해결하기 위해, Depth 카메라와 스테레오 카메라의 하이브리드 구성을 이용하여 정밀한 사용자 동공 추적 기법에 대해서 제안한다. 저해상도의 깊이 카메라의 얼굴 인식 정보를 바탕으로 고해상도 스테레오 카메라에서의 얼굴 및 눈의 후보영역을 찾고, 고해상도 스테레오 카메라에서 후보 영역 내의 동공 위치를 잦아서 빠르면서도 정밀한 동공 추적이 가능하도록 하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.11a
• /
• pp.1169-1172
• /
• 2010

대부분의 홍채인식 시스템의 전반부를 살펴보면 카메라를 통한 이미지를 획득 후 전처리 과정에서 동공의 경계를 잡는 과정이 수행된다. 카메라를 통한 이미지 획득 시 초점이 좋은 이미지와 나쁜 이미지가 같이 획득이 된다. 초점이 나쁜 이미지는 동공의 경계를 잡는 과정에 있어서 좋은 이미지보다 시간이 오래 걸리기 마련이다. 이에 본 논문에서는 영상획득 후 동공의 경계를 잡는 과정 전에 DCT(Discrete Cosine Transform)를 이용한 선명도 측정하는 방법을 제안한다. 카메라를 통한 영상을 획득한 후 초점이 좋은 영상과 나쁜 영상을 구분하게 되고 초점이 좋은 영상만을 선택하여 다음 과정인 동공의 경계를 잡는 작업을 수행하게 된다. 제안하는 방법은 DCT를 이용한 이미지의 선별 작업에 시간이 소비 되지만 나쁜 영상을 이용하여 동공의 경계를 잡느라 걸리는 시간을 줄일 수 있는 장점을 기대할 수 있다. 우선적으로 수학적 분석을 통하여 23%의 속도 절감을 통한 성능 개선의 가능성을 보였고, 실제 구현을 통하여 속도 향상이 기대된다.

• Journal of Korean Ophthalmic Optics Society
• /
• v.1 no.2
• /
• pp.7-11
• /
• 1996

The size of pupil with pupillary light reflex is determined by the amount of a luminance. and it is dependent with the distribution function of the retinal illuminance which is the amount of transmittance for the external light due to the size of pupil, and the detector of cone and rod due to the amount of the luminance. The change of the pupil size with the luminance can be expressed with the mathematical model $$y(x)={\alpha}+{\beta}\frac{1}{1+{e}{x}{p}(x-x_0)/{\theta}}$$ where ${\alpha}$ is the size of the pupil diameter in a maximum value of the luminance, ${\beta}$ is the deviation of the pupil's diameter between maximum and minimum, ${\theta}$ is the parameter showed the degree of a sensitivity. Comparing with the experimental value of P.Moon et al, We known that the equation of the model is very compatible.

• Korean Journal of Cognitive Science
• /
• v.14 no.3
• /
• pp.37-49
• /
• 2003

Most 3D display systems heavily depend on binocular disparity to produce 3-dimensional depth of a scene. In principle, the vergence angle of the object on fixation and binocular disparity of non-fixated objects vary with the inter-pupillary distance(IPD) of the observer. However, most stereo systems provide the identical stereo image pairs regardless of the observers' IPD, which may result in variation in the perceived depth. In this study, we manipulated the vergence angle of the fixated object and binocular disparity of the non-fixated object. The range of the individual difference in the perceived depth was found to be increased with the increase of disparity for both the fixated and non-fixated objects, and the individual difference was well fitted by the regression line of the observers' IPD. These results suggest that individual difference in the perceived depth from the identical stereo images should be greatly reduced if the stereo system calibrates the disparity of the object by the observers' IPD in generating the stereo images and the regression line found in this study might be useful in the calibrating the disparity of the images.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.5
• /
• pp.1162-1170
• /
• 2014

In this paper, an accurate remote gaze tracking method with two cameras is presented using a modified Starburst algorithm and honography normalization. Starburst algorithm, which was originally developed for head-mounted systems, often fails in detecting accurate pupil centers in remote tracking systems with a larger field of view due to lots of noises. A region of interest area for pupil is found using template matching, and then only within this area Starburst algorithm is applied to yield pupil boundary candidate points. These are used in improved RANSAC ellipse fitting to produce the pupil center. For gaze estimation robust to head movement, an improved homography normalization using four LEDs and calibration based on high order polynomials is proposed. Finally, it is shown that accuracy and robustness of the system is improved using two cameras rather than one camera.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.35 no.6
• /
• pp.539-547
• /
• 2007

In this paper, development and tests of a real-time eye-tracker system are discussed. The tracker system tracks a user's gaze point through movement of eyes by means of vision-based pupil detection. The vision-based method has an advantage of detecting the exact positions of user's eyes. An infrared camera and a LED are used to acquire a user's pupil image and to extract pupil region, which was hard to extract with software only, from the obtained image, respectively. We develop a pupil-tracking algorithm with Kalman filter and grab the pupil images by using DSP(Digital Signal Processing) system for real-time image processing technique. The real-time eye-tracker system tracks the movements of user's pupils to project their gaze point onto a background image.

• Korean Journal of Cognitive Science
• /
• v.14 no.1
• /
• pp.17-28
• /
• 2003

In order to provide natural images with a specified depth through three-dimensional display system, the stereo images should be similar to those projected from real environment as much as possible. Even when two persons see an identical scene, the binocular Parallax between two images of an object varies as a function of one's inter-pupilary distance (IPD). In this study, we investigated whether individual differences, such as IPD and accommodative vergence, would affect the perception of three dimensional scene provided by stereo-images. Results showed that a person's IPD is correlated with the limit of screen and binocular parallax for single vision, and affects the perceived depth of an object on fixation. More specifically, with longer IPD the limit of screen and binocular parallax for single vision is decreased, and the perceived depth is reduced. These results suggest that the screen and binocular parallax of an object should be calibrated with regard to users IPD to provide natural stereo-images with a specified depth and to Prevent double vision.

• Smart Media Journal
• /
• v.7 no.2
• /
• pp.15-22
• /
• 2018

Pupil tracking technologies can be used as an efficient information provider means that provides convenience to the user by connecting with a smart device. In this paper, we measure the distance of user gaze point using the pupil tracking device which produced by Pupil-labs, also shows the experimental result with analyzing accuracy and precision. Based on that the pupil gaze point location which tracked by pupil tracking device is compared with object target in terms of error. Since the mobile pupil tracking device is also one kind of camera, we have to perform the calibration before using the device. Not only generally used 2-dimensional calibration, but also 3-dimensional calibration method is explained. To get the improved accuracy of 2-dimensional calibration result, the 3-dimensional calibration set an imaginary plane and executes the calibration in various 3-dimensional spaces. To show the efficiency of 3-dimensional calibration, we analyze the experimental result. It also introduces various using methods and information that can be obtained through the device.

• Korean Journal of Forensic Psychology
• /
• v.11 no.2
• /
• pp.155-171
• /
• 2020

The purpose of this study is to examine the effects of emotional arousal and cognitive load on pupil diameter during a lie detection interview. The guilty group (n = 30) committed a mock crime (i.e., stealing cash) and the innocent group (n = 30) performed a mission (i.e., sending a message) in the research assistant's office. After that, their pupil size was measured using a wearable eye-tracker during the interview. The interview questions were classified with the three cognitive load, three emotional arousal, and three neutral questions. The results indicate that the main effects of group and time were not significant, but the interaction between group and time was significant. It means that when answering cognitive load questions, the guilty group showed larger increase in pupil diameter than the innocent group. The present study suggests that inducing cognitive load is more effective than inducing emotional arousal during an interview when using pupil diameter as an index of deception, and it is expected to improve the accuracy of lie detection.