퍼지 알고리즘을 이용한 관상동맥의 협착부위 검출

이주원,김성후,김주호,이한욱,정원근,이건기,Lee, Ju-Won,Kim, Sung-Hu,Kim, Joo-Ho,Lee, Han-Wook,Jung, Won-Geun,Lee, Gun-Ki 한국정보통신학회 2011 한국정보통신학회논문지 Vol.15 No.9

관상동맥 성형술과 관상동맥 우회술은 심근경색의 치료를 위해서 널리 이용되고 있는 방법이다. 특히 이 시술을 위해서 혈관의 협착 부위를 정확히 진단하는데 많은 어려움이 있다. 이러한 문제점을 개선하기 위해서 여러 연구자들에 의해서 에지 추출을 이용한 혈관의 협착부위를 검출하는 연구가 진행되고 있다. 그러나 이러한 방법들은 혈관의 구조나 영상의 질에 따라 그 성능의 차이가 발생한다. 따라서 본 연구는 이러한 문제점을 개선하기 위해 새로운 알고리즘을 제안한다. 제안된 알고리즘은 다중 샘플링과 쓰레쉬홀드, 퍼지 알고리즘을 이용한 혈관의 분기점과 끝점, 협착 부위를 검출하는 방법으로 구성되어 있다. 제안된 알고리즘의 성능을 평가하기 위해 다양한 혈관조영영상을 사용하였으며 그 결과 제안된 알고리즘에 의한 혈관의 분기점 및 끝점, 협착 부위 검출에 효과적이었다. Coronary angioplasty and coronary artery bypass graft, both are for the treatment of myocardial infarction widely used methods. For these procedures, there are especially difficulties in stenosis of blood vessels to diagnose accurately. To remedy this problem, by several researchers by using edge detection to detect stenosis of blood vessels has been studying. However, the results of using these methods vary defend on the vascular structure and the quality of the image. In this study, to improve these problems, the new algorithm is proposed. The proposed algorithm consists of methods to detect bifurcation of blood vessels and its ending point by using multi sampling, threshold and fuzzy algorithm. To evaluate the performance of the proposed algorithm, angiography was used for the different results of the blood vessels of the proposed algorithm, and the result was effective in detecting bifurcation of blood vessels and its ending point.

흉부 X-Ray 영상개선을 위한 신경망 적용에 관한 연구

이주원,이한욱,이종회,신태민,김영일,이건기,Lee, Ju-Won,Lee, Han-Wook,Lee, Jong-Hoe,Shin, Tae-Min,Kim Young-Il,Lee, Gun-Ki 대한전자공학회 2000 電子工學會論文誌-SC (System and control) Vol.37 No.1

흉부의 병변을 진단하기 위해 주로 사용되고 있는 흉부 X-선 촬영은 최근 컴퓨터 기술의 발달에 힘입어 디지털화 되고 있다. 디지털화된 흉부 영상을 방사선과 전문의가 모니터 상에서 관찰할 때 흉부 영상의 품질이 고르지 못하여 병변을 검출하기가 어려울 뿐만 아니라 이로 인하여 많은 진단 시간이 소요된다. 따라서 본 연구에서는 디지털 흉부 영상을 개선하기 위해 신경망을 이용하여 흉부 X-선 영상의 등화 방법을 제안하고 그 결과를 히스토그램 등화 방법과 비교하여 제시하였다. Recently, X-ray chest rediography is showing a tendency to take an image of digital radiography so as to diagnose the pathology of chest in a usual. When the radiologist observes the chest image derived from digital radiography system on the monitor, he feels difficult to find out the pathological pattern because the quality of chest radiography is unequal. It takes amount of time to adjust the proper image for diagnosis. Therefore, we propose the method of the chest image equalization using neural networks and provide the compared result with histogram equalization method.

콩 발아 중 isoflavone malonylglucosides의 함량 변이

이주원,이유정,이주희,조민식,최도진,마무현,김홍식,김대옥,윤홍태,김용호,Lee, Ju-Won,Yi, Yoo-Jung,Lee, Ju-Hee,Jo, Min-Sik,Choi, Do-Jin,Ma, Mu-Hyun,Kim, Hong-Sik,Kim, Dae-Ok,Yun, Hong-Tae,Kim, Yong-Ho 한국작물학회 2018 한국작물학회지 Vol.63 No.3

Soybean produces three major types of isoflavones, daidzein, genistein, and glycitein aglycones and their glucosides and malonylglucosides. It has been known that malonylated glucosides are rapidly converted to their corresponding aglycones due to the unstable thermolabile glucoside malonates; therefore, the analytical study of malonylated glucosides has been insufficient. In this study, we analyzed the malonylglucoside content in soybean seeds. Isoflavone analysis of three soybean cultivars revealed that 81.5~90.0% of the total isoflavones were malonylglucosides, whereas aglycones were rarely detected. Moreover, the total isoflavone content increased during a 5-day germination period where growth regulators and coumaric acid treatments tended to yield higher isoflavone content than the normal germination treatment, however the differences were not significant; notably, the isoflavone accumulation trend continued with additional germination days. The content of malonylglucoside was higher than that of other isoflavones, which was 83.7~86.6% of the total isoflavone content in seeds with a 3-day germination period. Furthermore, isoflavones were significantly accumulated in the hypocotyl of seedlings with a 5-day germination period. The content of isoflavone in the hypocotyl of the Pungsannamul-kong was 10,240 ug/g when treated with coumaric acid, which was considerably higher than that of other cultivars and treatments. Additionally, soybean seeds heated at $60^{\circ}C$ for 1 hour produced higher isoflavone content than non-heated soybean seeds. Our results show that it is possible to increase the isoflavone content in soybean seeds through various treatments.

Plate theory 를 이용한 Deoxyribonucleosides 의 분배 계수의 계산

이주원,소명섭,정성택,노경호 ( Ju Weon Lee,Myong Seup So,Sung Taik Chung,Kyung Ho Row ) 한국화학공학회 1996 화학공학의이론과응용 Vol.2 No.2

지정맥 인식을 위한 ROI 검출과 정맥 증강처리

이주원,이병로,Lee, Ju-Won,Lee, Byeong-Ro 한국정보통신학회 2015 한국정보통신학회논문지 Vol.19 No.4

최근, 생체 정보를 이용하여 개인의 신원을 확인하기 위해 근적외선 LED와 CCD 카메라를 이용한 지정맥 인식기술 연구되고 있다. 지정맥 인식은 손가락의 두께, 주변 광, 체온카메라 등의 잡음으로부터 정맥과 배경 이미지를 분리하는 방법에 따라 성능의 차이가 발생한다. 이러한 문제점을 개선하기 위해 본 연구에서는 NIR LED와 CCD 카메라로 촬영된 지정맥 영상으로부터 지정맥 회전, ROI 검출, 필터뱅크를 이용한 정맥 증강 기법을 제안하고 실험을 통하여 그 성능을 평가하였다. 이 실험의 결과에서 제안된 지정맥 회전과 ROI 검출의 정확도가 99.8%를 보였다. 그리고 필터뱅크를 이용한 정맥 증강처리에서는 제안된 방법이Retinex 알고리즘 보다 우수한 대비 성능을 보였다. 이 실험의 결과로부터 제안된 방법을 정맥인식의 전처리 과정에 적용한다면, 보다 나은 인식률을 제공할 것으로 사료된다. Recently, the finger vein recognition based on NIR and CCD sensor camera is investigating the technology to identify a personal using by biometrics. The performance difference of finger vein recognition is generated according to methods that are to separate the vein and background from noises such as finger thickness, ambient light, skin temperature, etc. To improve these problems, in this study, we are proposing the methods for rotation, ROI extraction, and enhancement of vein image captured by NIR LED and CCD camera, and were evaluated performances of these methods. In results of the experiment, the accuracy of the proposed method for image rotation and ROI extraction was 99.8%. And the proposed filter bank method in vein enhancement has shown better performance than retinex algorithm. The proposed method for results of these experimentations will provide better recognition rate when applied to the preprocessing of finger vein recognition.

우리나라 일본뇌염유행(日本腦炎流行)의 추이(趨移) -과거(過去) 20년간(年間)을 중심(中心)으로-

이주원,오대규,Lee, Chu-Won,Oh, Dae-Kyu 대한예방의학회 1987 예방의학회지 Vol.20 No.1

The following facts have been identified as a result of epidemiological trend and characteristic of Japanese Encephalitis in Korea for the last 20 years. First: The Epidemiological period which was ten-year and three-year in the past has been disappeared following the start of immunization program at 1970. Second: The Incidence rate was much higher in the south and West areas than northeast area of Korea. City and Province with the highest incidence rate was Chungcheong Nam Province and Cholla Buk Province. Third: Regardless of scope of prevalence, the main season that 90 percent of total incidence occurrs in one month from mid-August through mid-September. Fourth: The number of case by age was that 80 percent of total patients is children aged $3{\sim}15$. Recently there is an increase in the number of patients who are elderly people. Fifth: The study on the ecological conditions of mosquito including wintering and effectiveness of immunization for Japanese Encephalitis and duration on antibody should be done. Sixth: There has been no case of Japanese Encephalitis for the last three years since 1984 mainly due to disinfecting to eradicate mosquitos, immunization for vulnerable group of people aged $3{\sim}15$, individual precaution not to be bitten by mosquito, improvement of environment sanitation. While there has been no case of Japanese Encephalitis during last three years, there is possibility that Japanese Encephalitis becomes prevalent again anytime since its virus has been isolated continuously from the natural reservoirs.

원격 카메라 로봇 제어를 위한 동적 제스처 인식

이주원,이병로,Lee Ju-Won,Lee Byung-Ro 한국정보통신학회 2004 한국정보통신학회논문지 Vol.8 No.7

본 연구에서는 원격 카메라 로봇 제어를 위한 새로운 제스처 인식 방법을 제안하였다. 제스처 인식의 전처리 단계인 동적 제스처의 세그먼테이션이며, 이를 위한 기존의 방법은 인식 대상에 대한 많은 칼라정보를 필요로 하고, 인식단계에서는 각각 제스처에 대한 많은 특징벡터들을 요구하는 단점이 있다. 이러한 단점을 개선하기 위해, 본 연구에서는 동적 제스처의 세그먼테이션을 위한 새로운 Max-Min 탐색법과 제스처 특징 추출을 위한 평균 공간 사상법과 무게중심법, 그리고 인식을 위한 다층 퍼셉트론 신경망의 구조 둥을 제안하였다 실험에서 제안된 기법의 인식율이 90%이상으로 나타났으며, 이 결과는 원격 로봇 제어를 위한 휴먼컴퓨터 인터페이스(HCI : Human Compute. Interface)장치로 사용 가능함을 보였다. This study is proposed the novel gesture recognition method for the remote camera robot control. To recognize the dynamics gesture, the preprocessing step is the image segmentation. The conventional methods for the effectively object segmentation has need a lot of the cole. information about the object(hand) image. And these methods in the recognition step have need a lot of the features with the each object. To improve the problems of the conventional methods, this study proposed the novel method to recognize the dynamic hand gesture such as the MMS(Max-Min Search) method to segment the object image, MSM(Mean Space Mapping) method and COG(Conte. Of Gravity) method to extract the features of image, and the structure of recognition MLPNN(Multi Layer Perceptron Neural Network) to recognize the dynamic gestures. In the results of experiment, the recognition rate of the proposed method appeared more than 90[%], and this result is shown that is available by HCI(Human Computer Interface) device for .emote robot control.

단이론과 모멘트방법을 이용한 데옥시리보뉴클레오사이드의 평형상수의 계산 및 비교

이주원,노경호 ( Ju Weon Lee,Kyung Ho Row ) 한국공업화학회 1997 공업화학 Vol.8 No.3

데옥시리보뉴클레오사이드의 평형상수를 단이론과 모멘트방법에 의해서 일정 용매 조성법 역상 고성능 액체크로마토그래피(Reversed-Phase High Peformance Liquid Chromatography)를 이용하여 계산하였다. 이동상은 물을 기본으로 유기용매로서 acetonitrile과 methanole을 사용하였다. 단이론은 선형 흡착식으로부터 나온 모델로 일정한 유속으로 용매가 colunm의 단을 평형을 이루면서 지나간다는 가정 하에서 이루어졌다. 모멘트방법은 실험결과로부터 얻은 peak에서 첫 번째 절대 모멘트를 구하여 평형상수를 계산하는 방법이다. 이 두 가지 방법에 의한 다섯 가지의 데옥시리보뉴클레오사이드의 평형상수값은 서로 근사하였고, 용량인자를 이용하여 구한 평형상수의 값과도 비슷하였다. 평형상수는 이동상에서의 organic modifier의 양의 semi-log의 관계식으로 표시되었고 단이론에 의해 계산된 용출곡선은 실험값과 잘 일치하였다. The equilibrium constants of five deoxyribonucleosides (dCyd, dUrd, dGuo, dThd, dAdo) were estimated by the plate theory and the moment method under isocratic conditions of the Reversed-Phase High Performance Liquid Chromatography (RP-HPLC). The mobile phase in this system was composed of water and organic modifiers(acetonitrile and methanol). The plate theory of linear adsorption isotherm was treated on the basis of continuous flow of eluent through the plates of the column. The moment method was utilized to find the equilibrium constant from the first absolute moment of experimental data. The equilibrium constants of five deoxyribonucleosides in the two methods were very close, and also the equilibrium constants calculated by capacity factor were similar to those by both the plate theory and the moment method. The equilibrium constant was expressed as a semi-log function of the quantity of organic modifier. Excellent agreements between the calculated elution profile by the plate theory and the experimental data were observed.

집합 유사 시퀀스 매칭의 성능 향상을 위한 인덱스 기반 검색 방법

이주원,임효상,Lee, Juwon,Lim, Hyo-Sang 한국정보처리학회 2017 정보처리학회논문지. 소프트웨어 및 데이터 공학 Vol.6 No.11

집합 유사 시퀀스 매칭 방법은 유사한 정도를 나타내는 척도로 교집합을 기반으로 한 유사도를 사용한다. 그러나 교집합 크기를 계산하는 과정에 시간이 오래 걸릴 뿐만 아니라, 유사한 시퀀스를 찾기 위해서 수많은 집합 간 교집합 크기를 구해야 하므로 수행 시간이 오래 걸리는 성능상의 문제가 있다. 본 논문에서는 이러한 성능상의 문제를 해결하기 위해 인덱스 기반의 검색 방법을 사용하여 집합 기반 유사 시퀀스 매칭을 빠르게 수행하는 방법을 제안한다. 제안하는 방법은 크게 두가지로 구분된다. 첫 번째로 집합 시퀀스 유사도 문제를 교집합의 크기 비교 문제로 정형적으로 변환하고, 교집합의 크기를 빠르게 찾을 수 있는 인덱스 구조를 제안한다. 두 번째로 제안한 인덱스 구조를 사용하여 집합 기반 유사 시퀀스 매칭을 효율적으로 수행할 수 있는 방법을 제안한다. 성능 평가 결과, 제안하는 방법이 기존 방법에 비해 최대 30배에서 50배의 수행 시간 단축이 있음을 보인다. 또한 데이터 시퀀스의 개수가 증가할수록 수행시간의 차이가 점점 커지므로, 대용량 데이터 처리에 적절함을 보인다. The set-based similar sequence matching method measures similarity not for an individual data item but for a set grouping multiple data items. In the method, the similarity of two sets is represented as the size of intersection between them. However, there is a critical performances issue for the method in twofold: 1) calculating intersection size is a time consuming process, and 2) the number of set pairs that should be calculated the intersection size is quite large. In this paper, we propose an index-based search method for improving performance of set-based similar sequence matching in order to solve these performance issues. Our method consists of two parts. In the first part, we convert the set similarity problem into the intersection size comparison problem, and then, provide an index structure that accelerates the intersection size calculation. Second, we propose an efficient set-based similar sequence matching method which exploits the proposed index structure. Through experiments, we show that the proposed method reduces the execution time by 30 to 50 times then the existing methods. We also show that the proposed method has scalability since the performance gap becomes larger as the number of data sequences increases.

격벽형 전자종이의 하전입자 필터링 방법 및 구동조건에 따른 응답시간 및 반사율 분석

이주원,김영조,Lee, Joo-Won,Kim, Young-Cho 한국전기전자재료학회 2020 전기전자재료학회논문지 Vol.33 No.6

For electronic paper displays using electrophoresis, the response time and reflectivity of the image panel fabricated by filtering are analyzed. For the filtering process, a square wave and ramp wave are applied to white charged particles with a unique q/m value. We divide the sample panels into #1 to #4 according to the applied waveform in the filtering process. Step waves comprising two steps are used to drive the panel; therefore, we divide the driving conditions into D1~D4. The applied voltage at the first stage of the half cycle of the driving waveform moves the charged particles attached via the image force from the electrode, and the applied voltage at the second stage moves the floating charged particles by detaching. As mentioned, four types of driving conditions (D1 to D4) classified according to the half cycle of the driving waveform are applied to the samples #1 to #4), which are classified according to four types of filtering process. When driving condition D1 is applied to the four types of sample panels, the rise time of #1 is 1.59s, #2 is 1.706s, #3 is 1.853s, and #4 is 1.235s, resulting in #4 being relatively faster compared with other sample panels, and showing the same trend in other driving conditions. As a result, we confirm that applying the driving condition D1 causes abrupt movement of the white charged particles injected into the cell. When the same driving waveform (D1) is applied to each sample, reflectivities of 32.1% for #1, 31.4% for #2, 27.9% for #3, and 63.4% for #4 are measured. From the experiment, we confirm that the driving condition D1 (1s of 3.5 V, 9s of 3.0 V) and ramp wave #4 in filtering are desirable for good reflectivity and response time. Our research is expected to contribute to the improvement of the filtering process and optimization of the driving waveform.