• Korean Journal of Cognitive Science
• /
• v.13 no.2
• /
• pp.37-45
• /
• 2002

Despite dynamical retinal image changes caused by pursuit eye movements, we usually perceive the stable spatial properties of the environment suite successfully Helmholtz and his followers have suggested that the visual system coordinates the retinal and extraretinal eye position information to represent the spatial properties of the environment. However. there have been a significant amount of researches showing that this kind of mechanism may not operate perfectly, and the pursuit eye movement employed in those researches were limited to conjugate eye movements. When an observer tracks an object moving away from the observer with his/her eyes. the two eyes rotate in opposite direction. and this kind of disjunctive eye movement may produce undesirable binocular disparities for the objects in the background. The present study examined whether the visual system compensated for the undesirable binocular disparities caused by disjunctive eye movements with extraretinal eye position information. Although the target object was presented frontoparellely to the subjects. the subjects reported that the object was slanted toward (or alway from) them in consistent with the undesirable binocular disparities produced by the disjunctive eye movements. These results imply that the visual system may not perfectly compensate for the undesirable binocular disparities with extraretinal eye position information.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.4
• /
• pp.249-254
• /
• 2007

This paper describes a technique for detecting the boundary of the optic disk in digital image of the retina using inward and outward curve evolution. Optic disk boundary offers medical information about glaucoma progresses. For accurate boundary detection, image inpainting based on PDE removes blood vessels crossing the optic disk. For removing noises and preserving boundary of optic disk in image inpainting process, the anisotropic diffusion filtering is developed. After pre-processing, the optic disk boundary is determined using inward and outward curve evolution. Experimental results show that blurring effect of original region and optic disk boundary is reduced considerably. By the proposed method, we can detect correct disk boundary compare to conventional method.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.372-375
• /
• 2003

The incidence of blindness resulting from diabetic retinopathy has significantly increased despite the intervention of insulin to control diabetes mellitus. Early signs are microaneurysms, exudates, intraretinal hemorrhages, cotton wool patches, microvascular abnormalities, and venous beading. Advanced stages include neovascularization, fibrous formations, preretinal and vitreous microhemorrhages, and retinal detachment. Microaneurysm count is important because it is an indicator of retinopathy progression. The purpose of this paper is to apply data mining to detect diabetic retinopathy patterns in routine fundus fluorescein angiography. Early symptoms are of principal interest and therefore the emphasis is on detecting microaneurysms rather than vessel tortuosity. The analysis does not involve image-recognition algorithms. Instead, mathematical filtering isolates microaneurysms, microhemorrhages, and exudates as objects of disconnected sets. A neural network is trained on their distribution to return fractal dimension. Hausdorff and box counting dimensions grade progression of the disease. The field is acquired on fluorescein angiography with resolution superior to color ophthalmoscopy, or on patterns produced by physical or mathematical simulations that model viscous fingering of water with additives percolated through porous media. A mathematical filter and neural network perform the screening process thereby eliminating the time consuming operation of determining fractal set dimension in every case.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.10a
• /
• pp.393-397
• /
• 2007

최근 시각 장애인을 위한 인공망막 모델 구현에 관한 연구 중 시피질 자극기 기술은 시각 자극 전달의 중간 단계를 생략하고 직접 뇌세포를 자극하는 것이다. 본 논문에서는 망막에서 시각 피질로 시각정보를 전달할 때 발생하는 시각 피질의 특성, 즉 방향성에 대한 반응 특성을 특징 데이터로 구성하여 인식함으로써 인간 시각 정보 처리와 유사한 영상 추출 및 인식 모델을 제안한다. 제안된 방법은 영상의 특징을 추출 한 후 Delta-bar-delta 기반 오류 역전파 알고리즘을 적용하여 영상의 특징들을 인식한다. 제시된 방법의 성능을 분석하기 위하여 다양한 숫자 패턴들을 대상으로 실험한 결과, 제안된 망막 세포로부터 전달된 정보를 방향성에 대한 민감성을 고려하여 영상의 특성을 추출하여 인식하는 모델이 기존의 영상 추출 및 인식 모델보다 인식률에 있어서는 별 차이가 없지만 다양한 실험에서 확인할 수 있듯이 인간 시각과 같이 인식 성능이 민감하지 않는 것을 알 수 있었다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.04a
• /
• pp.261-264
• /
• 2005

현재 인간시각의 모델에 관한 많은 연구가 진행중이다. 본 논문에서는 형태정보의 영역대비를 강조하는 무축삭세포의 기능을 반영한 계산론적 망막모델을 제안한다. 무축삭세포는 전달된 물체의 운동정보의 변화를 감지하는 기능을 가지며, 그 감지된 정보를 강조하는 기능이 있다. 본 논문에서는 양극세포에서 출력된 형태정보의 영역대비를 강조한 계산론적 망막모델을 구현하였다. 실험에서는 양극세포의 결과 영상과 무축삭세포의 기능을 처리한 결과영상을 비교하였다. 따라서 무축삭세포의 영역대비 기능을 구현함으로써 대뇌피질에서는 영상의 정보를 효율적으로 처리할 수 있다.

• Annual Conference of KIPS
• /
• 2000.04a
• /
• pp.902-906
• /
• 2000

망막 혈관 구조의 분석은 망막에 관련된 환자의 진단 및 치료에 중요한 정보를 제공하기 때문에 다양한 연구가 진행되어 왔다. 본 연구에서는 ICG(IndoCyanine Green) 기술을 이용한 망막 영상의 혈관 구조를 분석하기 위해 원의 방정식으로 묘사된 혈관 영역 에너지 함수와 분기점 추정 템플릿을 사용하여 혈관의 특징점들을 추정한 후 혈관의 형체(body)를 자동으로 추적하는 동시에 분기점을 추출하는 방법을 제안한다. 전체 혈관의 자동추적과 분기점 추출을 가능하게 하는 특징점 추정 방법과 혈관 형체의 자동추적 알고리즘 및 분기점 추출 방법을 ICG 망막 영상에 적용하여 실험한 결과 만족할 만한 성능을 보였다.

• International Journal of Computer Science & Network Security
• /
• v.24 no.9
• /
• pp.119-126
• /
• 2024

Diabetes is a chronic condition that happens when the pancreas fails to produce enough insulin or when the body's insulin is ineffectively used. Uncontrolled diabetes causes hyperglycaemia, or high blood sugar, which causes catastrophic damage to many of the body's systems, including the neurons and blood vessels, over time. The burden of disease on the global healthcare system is enormous. As a result, early diabetes diagnosis is critical in saving many lives. Current methods for determining whether a person has diabetes or is at risk of acquiring diabetes, on the other hand, rely heavily on clinical biomarkers. This research presents a unique deep learning architecture for predicting whether or not a person has diabetes and the severity levels of diabetes from the person's retinal image. This study incorporates datasets such as EyePACS and IDRID, which comprise Diabetic Retinopathy (DR) images and uses Dense-121 as the base due to its improved performance.

• Korean Journal of Cognitive Science
• /
• v.34 no.3
• /
• pp.181-196
• /
• 2023

Size, velocity, and time equivalence are mechanisms that allow us to perceive objects in three-dimensional space consistently, despite errors on the two-dimensional retinal image. These mechanisms work on common cues, suggesting that the perception of motion distance, motion speed, and motion time may share common processing. This can lead to the hypothesis that, despite the spatial nature of visual stimuli distorting temporal perception, the perception of motion speed and the perception of motion duration will tend to oppose each other, as observed for objects moving in the environment. To test this hypothesis, the present study measured perceived speed using Müller-Lyer illusion stimulus to determine the relationship between the time-perception consequences of motion stimuli observed in previous studies and the speed perception measured in the present study. Experiment 1 manipulated the perceived motion trajectory while controlling for the retinal motion trajectory, and Experiment 2 manipulated the retinal motion trajectory while controlling for the perceived motion trajectory. The result is that the speed of the inward stimulus, which is perceived to be shorter, is estimated to be higher than that of the outward stimulus, which is perceived to be longer than the actual distance traveled. Taken together with previous time perception findings, namely that time perception is expanded for outward stimuli and contracted for inward stimuli, this suggests that when the perceived trajectory of a stimulus manipulated by the Müller-Lyer illusion is controlled for, perceived speed decreases with increasing duration and increases with decreasing duration when the perceived distance of the stimulus is constant. This relationship suggests that the relationship between time and speed perceived by spatial cues corresponds to the properties of objects moving in the environment, i.e, an increase in time decreases speed and a decrease in time increases speed when distance remains the same.

• Journal of the Optical Society of Korea
• /
• v.17 no.1
• /
• pp.68-72
• /
• 2013

We developed an SD-OCT (Spectral Domain-Optical Coherence Tomography) system which uses a GPU (Graphics Processing Unit) for processing. The image size from the SD-OCT system is $1024{\times}512$ and the speed is 110 frame/sec in real-time. K-domain linearization, FFT (Fast Fourier Transform), and log scaling were included in the GPU processing. The signal processing speed was about 62 ms using a CPU (Central Processing Unit) and 1.6 ms using a GPU, which is 39 times faster. We performed an in-vivo retinal scan, and reconstructed a 3D visualization based on C-scan images. As a result, there were minimal motion artifacts and we confirmed that tomograms of blood vessels, the optic nerve, and the optic disk are clearly identified. According to the results of this study, this SD-OCT can be applied to real-time 3D display technology, particularly auxiliary instruments for eye operations in ophthalmology.

• Journal of Biomedical Engineering Research
• /
• v.40 no.2
• /
• pp.48-54
• /
• 2019

This paper proposed a method of validation data augmentation for improving the grading accuracy of diabetic macular edema (DME) using deep learning. The data augmentation technique is basically applied in order to secure diversity of data by transforming one image to several images through random translation, rotation, scaling and reflection in preparation of input data of the deep neural network (DNN). In this paper, we apply this technique in the validation process of the trained DNN, and improve the grading accuracy by combining the classification results of the augmented images. To verify the effectiveness, 1,200 retinal images of Messidor dataset was divided into training and validation data at the ratio 7:3. By applying random augmentation to 359 validation data, $1.61{\pm}0.55%$ accuracy improvement was achieved in the case of six times augmentation (N=6). This simple method has shown that the accuracy can be improved in the N range from 2 to 6 with the correlation coefficient of 0.5667. Therefore, it is expected to help improve the diagnostic accuracy of DME with the grading information provided by the proposed DNN.