• 전자공학회논문지B
• /
• 제29B권9호
• /
• pp.37-49
• /
• 1992

The standard definition of computational vision is a set of inverse problems of recovering surfaces from images. Thus the common characteristics of the most early vision problems are ill-posed. The main idea for solving ill-posed problems is to restrict the class of admissible solutions by introducing suitable a priori knowledge. Standard regurarization methods lead to satisfactory solutions of early vision problems but cannot deal effectively and directly with a few general problems, such as discontinuity and fusion of information from multiple modules. In this paper, we discuss limitations of standard regularization theory and present new stochastic method. We will outline a rigorous approach to overcome part of ill-posedness of image restoration, edge detection, and stereo vision problems, based on Bayes estimation and MRF(Markov random field) model, that effectively deals with the problems. This result makes one hope that this framework could be useful in the solution of other vision problems.

• 한국정밀공학회지
• /
• 제14권12호
• /
• pp.17-23
• /
• 1997

A 3-D vision system is developed to evaluate blanks' line up and monitor gap and thickness difference between blanks in tailored blank welding system. A structured lighting method is used for 3-D vision recognition. Images of sheared portion in blanks are irregular according to roughness of blank surface, shape of sheared geometry and blurring. It is difficult to get accurate and reliable informations in the case of using binary image processing or contour detection techniques in real time for such images. We propoe a new energy integration method robust to blurring and changes of illumination. The method is computationally simple, and uses feature restoration concept, different to another digital image restoration methods which aim image itself restoration and may be used in conventional applications using structured line lighting technique. Experimental results show this system measuring repeatability is .+-. pixel for gap and thickness difference in static and dynamic tests. The data are expected to be useful for preview gap control.

• 한국동물생명공학회지
• /
• 제38권1호
• /
• pp.2-9
• /
• 2023

Mammalian eyes have a limited ability to regenerate once neurons degenerate. This results in visual impairment that impacts the quality of life among adult populations as well as in young children leading to lifelong consequences. Various therapies are in development to restore vision, and these include gene therapy, stem cell therapy, in-vivo transdifferentiation, and transplantation of a patient's whole eye obtained from interspecies blastocyst complementation. This review discusses advances in the research as well as hurdles that need to be resolved to have a successful restoration of vision.

• 대한원격탐사학회지
• /
• 제23권5호
• /
• pp.439-445
• /
• 2007

Stereoscopic viewing is an efficient technique for not only computer vision but also remote sensing applications. Generally, stereo pair obtained at the same time is necessary for 3D viewing, but it is possible to synthesize a stereomate suitable for stereo view with a single image and disparity-map. There have been researches concerning the generation of the synthetic stereomate from remote sensing imagery. However it is hard to find researches concerning the restoration of occlusion in stereomate. In this paper, we generated synthetic stereomates from remote sensing images, focused on the occlusion restoration. In order to figure out proper restoration methods depending on the spatial resolution of remote sensing imagery, we tested several methods including general interpolation and inpainting technique, then evaluated the results.

• 융합신호처리학회논문지
• /
• 제6권3호
• /
• pp.112-119
• /
• 2005

본 논문에서는 심리학 생리학적 지식을 기초로 하여 좌우 망막을 양안 입체시로 대응시켜 얻은 2매의 화상으로부터 깊이 정보를 추출하는 신경회로 모델을 제안하고 3차원 표면의 복원법을 검토한다. 화상의 특징을 근거로 하여 시차를 추출할 경우, 경계 부분에 유사한 특징이 반복된다면 경우 올바른 깊이 정보를 검출할 수 없다. 본 논문에서 제안된 신경회로 모델은 시차의 추출, 시차의 통합, 시차의 보간에 의하여 시차를 결정한다. 또한, 깊이 정보를 보간하여 3차원 형상을 복원하고 그 복원된 3차원 형상에 좌입력 화상을 투영하여 3차원 표면을 복원하는 법을 제안하고, 실험을 통하여 시차 추출 시간을 대폭 줄일 수 있음을 확인하였다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2013년도 춘계학술대회 논문집
• /
• pp.979-980
• /
• 2013

• Archives of Plastic Surgery
• /
• 제34권5호
• /
• pp.663-666
• /
• 2007

Purpose: Autologous fat injection into the facial area is a frequently used technique in aesthetic plastic surgery for augmentation of the soft tissue. Fat injection is a very safe procedure because of the advantage of being autologous tissue. Minimal foreign body reaction or infections are noted after fat injection. However, there may be some complications including those as severe as blindness. There have been some case reports on visual disturbances after autologous fat injection reported in the literature. Methods: A 21-year-old female patient underwent autologous fat injection into left eyebrow area to correct depression of soft tissue. Immediately after injection of autologous fat, she complained sudden visual loss on the left eye. She had come to our emergency room and ophthalmologic evaluation showed that the patient could only recognize hand motion. There was no abnormality of the optic nerve on magnetic resonance imaging. Suspecting an ischemic optic neuritis from fat embolism of the central retinal artery, the patient was treated conservatively with occular massage, antiglaucomatic agent, anti-inflammatory drugs and antibiotics. Visual field examination showed visual defect of half the lower hemisphere. Results: While maintaining antiglaucomatic agents and non steroidal anti inflammatory drugs, fundoscopic examination showed no abnormalities on the second day of admission. Visual field examination showed an improvement on the fourth day along with decreased eyeball pain. Significant improvement of vision was noted and the patient was discharged on the fifth day of admission. The patient was followed-up 2 days afterwards with improved vision and visual field defect. Conclusion: We describe an unusual case of sudden unilateral visual disturbance following autologous fat injection into periorbital area.

• 방송공학회논문지
• /
• 제23권4호
• /
• pp.503-510
• /
• 2018

손상된 영상의 복원은 디지털 영상 처리기술이 등장하기 이전부터 시도되었던 근원적 문제이다. 컴퓨터의 연산 능력과 다양한 기술의 발전에 따라 손상된 영상을 복원하는 다양한 연구가 소개되었으나 그 결과는 사람에 의한 수동적 결과물과 비교하여 낮은 복원 결과를 보여 왔다. 최근 심층 신경망 (DNN, Deep Neural Network)의 발전으로 이미지 복원에 이를 적용한 다양한 연구가 소개 되고 있지만, 광범위한 영역이 손상된 경우 근접한 화소를 활용하는 방법으로 해결이 어렵다. 이와 같은 경우는 주변의 영상의 문맥적 정보를 통해 손상된 영역을 추론을 통한 복원이 필요하다. 본 논문에서는 심층 신경망 기술 중 하나인 적대적 생성신경망(GAN, Generative Adversarial Network)을 이용한 이미지 복원 네트워크를 제안한다. 제안하는 시스템은 이미지 생성 네트워크, 생성 결과 판별 네트워크로 구성 된다. 본 논문에서는 제안하는 방안을 통해 다양한 종류의 이미지를 복원함에 있어서 훼손된 영역의 추론을 통하여 자연스러운 영상 복원뿐 아니라 원본 영상의 질감까지 복원이 가능함을 실험을 통해 확인 하였다.

• 한국컴퓨터정보학회논문지
• /
• 제19권4호
• /
• pp.9-16
• /
• 2014

본 논문은 화소 데이터의 비교를 이용한 단일 프레임 또는 연속 프레임에 나타나는 선형 스크래치를 탐지하여 복원하는 알고리즘을 제안하였다. 스크래치탐지와 복원방법은 프레임 간 많은 비교 연산시간을 필요로 하며 병렬처리 가능성이 높다. 제안하는 스크래치 탐지와 복원방법은 빠른 처리를 위해 GPU에서 수행할 수 있도록 병렬 설계 하였다. 제안하는 알고리즘은 국가 기록원 디지털화 영상에 대해 순차처리와 병렬처리의 성능 테스트를 수행하였다. 실험에서 연속한 스크래치를 고려하는 경우의 탐지율은 단일 프레임만 고려하는 방법보다 20% 이상 성능이 향상되었다. GPU 기반 알고리즘의 탐지율과 복원율은 CPU 기반의 알고리즘과 유사하였으나 50배 이상의 연산속도가 향상되었다.

• 한국멀티미디어학회논문지
• /
• 제25권5호
• /
• pp.676-684
• /
• 2022

As realistic media are widespread in various image processing areas, image or video compression is one of the key technologies to enable real-time applications with limited network bandwidth. Generally, image or video compression cause the unnecessary compression artifacts, such as blocking artifacts and ringing effects. In this study, we propose a Deep Residual Channel-attention Network, so called DRCAN, which consists of an input layer, a feature extractor and an output layer. Experimental results showed that the proposed DRCAN can reduced the total memory size and the inference time by as low as 47% and 59%, respectively. In addition, DRCAN can achieve a better peak signal-to-noise ratio and structural similarity index measure for compressed images compared to the previous methods.