• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.48 no.1
• /
• pp.24-30
• /
• 2011

We have developed a hand-held probe for an ophthalmic OCT system. The hand-held probe for imaging was designed to be compact and portable. The cornea and retinal images were acquired by replacing the objective lens at the front of the probe. To verify the performance of the hand-held OCT probe, we acquired two dimensional OCT image of the rat eye in vivo and reconstructed three dimensional rat eye rendering images. In vivo 3D OCT images were showed distinct structural information in the posterior and anterior chamber with minimal motion artifacts. Thereby, OCT imaging speed is suitable for an dynamic in vivo experiment.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.12
• /
• pp.672-678
• /
• 2020

Ultra-high-speed images that accurately depict the minute movements of objects have become common as low-cost and high-performance cameras that can film at high speeds have emerged. In this paper, the proposed method removes unexpected noise contained in images after input at high speed, and then extracts an area of interest that can represent personal information, such as skin areas, from the image in which noise has been removed. In this paper, noise generated by abnormal electrical signals is removed by applying bilateral filters. A color model created through pre-learning is then used to extract the area of interest that represents the personal information contained within the image. Experimental results show that the introduced algorithms remove noise from high-speed images and then extract the area of interest robustly. The approach presented in this paper is expected to be useful in various applications related to computer vision, such as image preprocessing, noise elimination, tracking and monitoring of target areas, etc.

• Journal of the Korean Society of Radiology
• /
• v.17 no.6
• /
• pp.939-945
• /
• 2023

This study analyzes the 'performance efficiency' of AI-based reading assistance systems in the field of radiology by measuring their 'time behavior' properties. Due to the increase in medical images and the limited number of radiologists, the adoption of AI-based solutions is escalating, stimulating a multitude of studies in this area. Contrary to the majority of past research which centered on AI's diagnostic precision, this study underlines the significance of time behavior. Using 50 chest X-ray PA images, the system processed images in an average of 15.24 seconds, demonstrating high consistency and reliability, which is on par with leading global AI platforms, suggesting the potential for significant improvements in radiology workflow efficiency. We expect AI technology to play a large role in the field of radiology and help improve overall healthcare quality and efficiency.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.10a
• /
• pp.431-435
• /
• 2008

2006년 11월 이후 신 차량 번호판 등장 후, 신 차량 번호판과 구 차량 번호판이 혼합되어 있다. 이에 따라 속도위반, 신호위반 단속, 무인 주차관리 시스템, 범죄 및 도주 차량 검거, 고속도로 톨게이트에서 통행료 지불로 인한 교통 체증현상을 해소하기 위한 자동 요금 징수와 같은 다양한 경우에서 자동차 번호판의 특징에 맞는 인식 시스템이 요구되고 있다. 따라서 본 논문에서는 이러한 문제를 해결하기 위해 형태학적 특징 및 차 연산과 ART2 알고리즘을 이용한 차량 번호판 인식 방법을 제안한다. 무인 카메라에서 획득된 차량 번호판 영상에서 차 연산을 이용하여 에지를 추출한 후에 블록 이진화를 한다. 이진화 된 차량 영상에서 신 구 차량 번호판의 형태학적 특성을 8방향 윤곽선 추적 알고리즘에 적용하여 잡음 영역을 제거하고, 차량의 번호판 영역을 추출한다 추출된 번호판 영역에 대하여 평균 이진화와 최대 최소 이진화를 적용하여 번호판의 개별 영역에 대한 형태학적 특성을 고려하여 잡음을 제거하고, Labeling 알고리즘을 적용하여 개별 문자를 추출한 후에 결합한다. 이렇게 분류된 개별 문자 및 숫자 코드를 ART2 알고리즘에 적용하여 학습 및 인식을 한다. 제안된 차량 번호판 추출 및 인식 방법의 성능을 평가하기 위해 녹색 번호판과 흰색 번호판 이미지 각각 100장을 대상으로 실험한 결과, 제시 된 차량 번호판 추출 및 인식 방법이 실험을 통해서 효율적인 것을 확인하였다.

• Journal of Broadcast Engineering
• /
• v.10 no.3
• /
• pp.383-391
• /
• 2005

In this paper, a new anisotropic diffusion based on Alvarez, Lions, and Morel (ALM) diffusion model is proposed for the suppression of blocking artifact caused by discrete cosine transform (DCT) based image compression. The proposed diffusion model, which incorporates a 'rate control parameter' (RCP), makes it possible to reduce blocking artifacts while to preserve the edge. The RCP controls the rate between isotropic and anisotropic diffusion. Isotropic diffusion is encouraged to eliminate the blocking artifacts in a block boundary of a smooth region, while anisotropic diffusion is encouraged to keep the edge or texture sharp in edge and a block boundary within an edge region. Additionally, to avoid oversmoothness of the texture region, a 'speed control parameter' (SCP), which makes diffusion process slow in the texture region, is employed.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.22 no.4
• /
• pp.386-392
• /
• 2002

Ultrasonic has been proposed as an attractive means of detecting bone loss. There have been several commercial ultrasound devices developed for measuring the heel to predict fracture at other bones. However, these devices select only single point of heel bone as measurement site. It causes poor assessment of bone quality due to the error of transducer positioning. In an effort to improve current ultrasound systems, we evaluated the linear scanning method which provides better prediction of bone quality and an accurate image of bone shape. The system used in this study biaxially scans a heel bone using automated linear scanning technique. The results demonstrated that the values of ultrasound parameters varied with different positions within bone specimen. It has been also found that the linear scanning method could better pre야ct bone quality, eliminating the error of transducer positioning.

• Journal of KIISE:Software and Applications
• /
• v.33 no.4
• /
• pp.411-417
• /
• 2006

Motion Estimation (ME) is an important part of video coding process and it takes the largest amount of computation in video compression. Half-pixel and quarter-pixel motion estimation can improve the video compression rate at the cost of higher computational complexity In this paper, we suggest a new efficient low-complexity algorithm for half-pixel and quarter pixel motion estimation. It is based on the experimental results that the sum of absolute differences(SAD) shows parabolic shape and thus can be approximated by using interpolation techniques. The sub-pixel motion vector is searched from the minimum SAD integer-pixel motion vector. The sub-pixel search direction is determined toward the neighboring pixel with the lowest SAD among 8 neighbors. Experimental results show that more than 20% reduction in computation time can be achieved without affecting the quality of video.

• Progress in Medical Physics
• /
• v.6 no.2
• /
• pp.83-92
• /
• 1995

The T1 mapping of an human anatomy may give a characteristic contrast among the various tissues and the normal/abnormal tissues. Here, the methodology of constructing T1 map out of several images with different TRs, will be described using non-linear curve fitting. The general curve fitting algorithm requires the initial trial values T1t and Mot for the variables to be fitted. Three different methods of suppling the trial T1t and Mot are suggested and compared for the efficiency and the accuracy. The curve-fitting routine was written in ANSI C and excuted on a SUN workstation. Several distilled-water phantoms with various concentrations of Gd-DTPA were prepared to examine the accuracy of the curve-fitting program. An MR image was used as the true proton density image without any random noise, and several images with different TRs were generated with the theoretical T1 relaxation times 250, 500, and 1000msec. The random noise of 1, 5, and 10％ were embedded into the simulated images. These images were used to generate the T1 map, and the resultant T1 maps for each T1 were analyzed to study the effect of the random noise on the T1 map.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.18 no.1
• /
• pp.166-173
• /
• 2014

A noncontact nondestructive testing (NDT) method is proposed to detect the damage of pipeline structures and to identify the location of the damage. To achieve this goal, a scanning laser source actuation technique is utilized to generate a guided wave and scans a specific area to find damage location more precisely. The ND: YAG pulsed laser is used to generate Lamb wave and a piezoelectric sensor is installed to measure the structural responses. The measured responses are analyzed using three dimensional Fourier transformation (3DFT). The damage-sensitive features are extracted by wavenumber filtering based on the 3D FT. Then, flaw imaging techniques of a pipeline structures is conducted using the damage-sensitive features. Finally, the pipes with notches are investigated to verify the effectiveness and the robustness of the proposed NDT approach.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.441-447
• /
• 2023

Intravenous infusion therapy is a standard nursing procedure in medical institutions that provides patients with drugs, fluids, blood, and nutrients into the patient's mucus. It is mainly performed and managed by nurses. Additionally, it is an injection method that injects drugs directly into the blood vessels, and is used to achieve rapid results in emergency situations, and quick and accurate effects can be expected. Even experienced nurses through education and training often make mistakes, which can not only cause discomfort to patients but also cause various problems that threaten patient safety. Various studies are being conducted to reduce the pain caused by these mistakes. This paper acquired images of veins on the back of the hands of three subjects through an image detection device and conducted a study to derive an algorithm to provide clear vein images through image processing of the collected images. To sharpen the acquired vein images, existing algorithms Histogram Equalization, CLAHE, and Unsharp Masking were selected and combined. A histogram graph was used to compare images derived by applying individual algorithms and algorithm combinations to images. The histogram graph was checked by calculating the difference between the minimum and maximum values of distributed pixels and averaging them. The algorithm combination presented in this paper was 209.1, which was higher than the average values of individual algorithms of 138.7, 132.3, and 126.2, and it was confirmed that visibility was good even in actual images.