• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.29-37
• /
• 2021

In this paper, we propose a deep learning-based retinal vessel segmentation model for handling multi-scale information of fundus images. we integrate the selective kernel convolution into U-Net-based convolutional neural network. The proposed model extracts and segment features information with various shapes and sizes of retinal blood vessels, which is important information for diagnosing eye-related diseases from fundus images. The proposed model consists of standard convolutions and selective kernel convolutions. While the standard convolutional layer extracts information through the same size kernel size, The selective kernel convolution extracts information from branches with various kernel sizes and combines them by adaptively adjusting them through split-attention. To evaluate the performance of the proposed model, we used the DRIVE and CHASE DB1 datasets and the proposed model showed F1 score of 82.91% and 81.71% on both datasets respectively, confirming that the proposed model is effective in segmenting retinal blood vessels.

• Analytical Science and Technology
• /
• v.24 no.6
• /
• pp.533-543
• /
• 2011

The hERG (human ether-a-go-go related gene) ion channel is a main factor for cardiac repolarization, and the blockade of this channel could induce arrhythmia and sudden death. Therefore, potential hERG ion channel inhibitors are now a primary concern in the drug discovery process, and lots of efforts are focused on the minimizing the cardiotoxic side effect. In this study, $IC_{50}$ data of 202 organic compounds in HEK (human embryonic kidney) cell from literatures were used to develop predictive 2D-QSAR model. Multiple linear regression (MLR), Support Vector Machine (SVM), and artificial neural network (ANN) were utilized to predict inhibition concentration of hERG ion channel as machine learning methods. Population based-forward selection method with cross-validation procedure was combined with each learning method and used to select best subset descriptors for each learning algorithm. The best model was ANN model based on 14 descriptors ($R^2_{CV}$=0.617, RMSECV=0.762, MAECV=0.583) and the MLR model could describe the structural characteristics of inhibitors and interaction with hERG receptors. The validation of QSAR models was evaluated through the 5-fold cross-validation and Y-scrambling test.

• The KIPS Transactions:PartA
• /
• v.14A no.2
• /
• pp.77-84
• /
• 2007

The intima media thickness(IMT) of the carotid artery from B mode ultrasound images has recently been proposed as the most useful index of individual atherosclerosis and can be used to predict major cardiovascular events. Ultrasonic measurements of the IMT are conventionally obtained by manually tracing interfaces between tissue layers. The drawbacks of this method are the inter and intra observer variability and its inefficiency. In this paper, we present a multiresolution snake method combined with the dynamic programming, which overcomes the various noises and sensitivity to initialization of conventional snake. First, an image pyramid is constructed using the Gaussian pyramid that maintains global edge information with smoothing in the images, and then the boundaries are automatically detected in the lowest resolution level by minimizing a cost function based on dynamic programming. The cost function includes cost terms which are representing image features and geometrical continuity of the vessel interfaces. Since the detected boundaries are selected as initial contour of the snake for the next level, this automated approach solves the problem of the initialization. Moreover, the proposed snake improves the problem of converging th the local minima by defining the external energy based on multiple image features. In this paper, our method has been validated by computing the correlation between manual and automatic measurements. This automated detection method has obtained more accurate and reproducible results than conventional edge detection by considering multiple image features.

• Journal of KIISE:Software and Applications
• /
• v.31 no.9
• /
• pp.1226-1232
• /
• 2004

Biometric authentication is rising technology for the security market of the next generation. But most of biometric systems are developed using only one of various biological features. Recently, there is a vigorous research for the standardization of various biometric systems. In this paper, we propose a web-based authentication system using three other verifiers based on functional, parametric, and structural approaches for one biometrics of handwritten signature, which is conformable to a specification of BioAPI introduced by BioAPI Consortium for a standardization of biometric technology. This system is developed with a client-server structure, and clients and servers consist of three layers according to the BioAPI structure. The proposed neb-based multiple authentication system of one biometrics can be used to highly increase confidence degree of authentication without additional several biological measurements, although rejection rate is a little increased. That is, the false accept rate(FAR) decreases on the scale of about 1:40,000, although false reject rate(FRR) increases about 2.7 times in the case of combining above three signature verifiers. So the proposed approach can be used as an effective identification method on the internet of an open network. Also, it can be easily extended to a security system using multimodal biometrics.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05c
• /
• pp.1737-1740
• /
• 2003

최근 몇 년간 웹 애플리케이션의 개발은 다른 애플리케이션의 개발보다 사회적으로 더욱 활발하였다. 이것은 우리 사회에 많이 보급되어 손쉽게 접근할 수 있는 인터넷과 또 그 인터넷 안에 있는 웹의 편리함에 그 이유가 있을 것이다. 그러나 이러한 활발한 웹 애플리케이션 개발에 비하여 웹 애플리케이션의 개발 방법 자체에 대한 연구는 거의 전무한 상태이다. 따라서 대학 교육에서도 그 일정한 지침이 없는 상태이다. 이러한 개발 방법에 대한 미진한 연구는 아마도 다른 애플리케이션에 비하여 웹 애플리케이션의 개발이 시작된 지 오래되지 않은 원인도 있겠지만 또 웹 애플리케이션이 가지는 특성에도 기인할 것이다. 즉, 웹은 보통 MVC(Model-View-Controller) 모델의 뷰(view)가 중심이 되고 컨트롤 부분이나 모델 부분은 상대적으로 적기 때문이다. 그래서 그 구현이 다른 일반 애플리케이션의 구현보다 비교적 개념이 쉽고 간단하다. 그러나 그 구현이 비교적 간단하다 하더라도 정형화된 개발 모델이 있으면 애플리케이션의 생산성은 향상될 뿐만이 아니라 학교 교육에도 큰 도움이 될 것이다. 즉 학생들에게 정형화 된 모델을 교육하는 경우 그것은 학생들의 웹 애플리케이션 개발에 큰 도움이 되는 것이다. 실제로 개발 모델을 교육시키고 프로젝트를 하는 경우와 그렇지 않은 경우에 그 완성도에서 크나큰 차이를 보였다. 정형화된 모델을 교육받은 학생들은 그렇지 않은 학생들에 비하여 프로젝트의 완성도도 높을 뿐만 아니라 그 구성이나 일관성에서 훨씬 좋은 결과를 보였다. 본 논문에서는 이러한 웹 애플리케이션을 위한 개발 방법을 제시하고 그에 앞서 클라이언트-서버 모델이 웹이라는 환경에서 가질 수 있는 특성과 작업에 대하여 분석한다. 그리고 본 논문에서는 그 예를 JSP로 보여준다.라도 단순 다중 쓰레드 모델보다 더 많은 수의 클라이언트를 수용할 수 있는 장점이 있다. 이러한 결과를 바탕으로 본 연구팀에서 수행중인 MoIM-Messge서버의 네트워크 모듈로 다중 쓰레드 소켓폴링 모델을 적용하였다.n rate compared with conventional face recognition algorithms. 아니라 실내에서도 발생하고 있었다. 정량한 8개 화합물 각각과 총 휘발성 유기화합물의 스피어만 상관계수는 벤젠을 제외하고는 모두 유의하였다. 이중 톨루엔과 크실렌은 총 휘발성 유기화합물과 좋은 상관성 (톨루엔 0.76, 크실렌, 0.87)을 나타내었다. 이 연구는 톨루엔과 크실렌이 총 휘발성 유기화합물의 좋은 지표를 사용될 있고, 톨루엔, 에틸벤젠, 크실렌 등 많은 휘발성 유기화합물의 발생원은 실외뿐 아니라 실내에도 있음을 나타내고 있다.>10)의 $[^{18}F]F_2$를 얻었다. 결론: $^{18}O(p,n)^{18}F$ 핵반응을 이용하여 친전자성 방사성동위원소 $[^{18}F]F_2$를 생산하였다. 표적 챔버는 알루미늄으로 제작하였으며 본 연구에서 연구된 $[^{18}F]F_2$가스는 친핵성 치환반응으로 방사성동위원소를 도입하기 어려운 다양한 방사성의 약품개발에 유용하게 이용될 수 있을 것이다.었으나 움직임 보정 후 영상을 이용하여 비교한 경우, 결합능 변화가 선조체 영역에서 국한되어 나타나며 그 유의성이 움직임 보정 전에 비하여 낮음을 알 수 있었다. 결론: 뇌활성화 과제 수행시에 동반되는 피험자의 머리 움직임에 의하여 도파민 유리가 과대평가되었으며 이는 이 연구에서 제안한 영상정합을 이용한 움직임 보정기법에 의해서 개선되었다. 답이 없는 문제, 문제 만들기

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1901-1910
• /
• 2022

Arriving in the new space age, securing technology for fusion application of KOMPSAT-3·3A and global satellite images is becoming more important. In general, multi-sensor satellite images have relative geometric errors due to various external factors at the time of acquisition, degrading the quality of the satellite image outputs. Therefore, we propose a fine-image registration methodology to minimize the relative geometric error between KOMPSAT-3·3A and global satellite images. After selecting the overlapping area between the KOMPSAT-3·3A and foreign satellite images, the spatial resolution between the two images is unified. Subsequently, tie-points are extracted using a hybrid matching method in which feature- and area-based matching methods are combined. Then, fine-image registration is performed through iterative registration based on pyramid images. To evaluate the performance and accuracy of the proposed method, we used KOMPSAT-3·3A, Sentinel-2A, and PlanetScope satellite images acquired over Daejeon city, South Korea. As a result, the average RMSE of the accuracy of the proposed method was derived as 1.2 and 3.59 pixels in Sentinel-2A and PlanetScope images, respectively. Consequently, it is considered that fine-image registration between multi-sensor satellite images can be effectively performed using the proposed method.

• Proceedings of the ESK Conference
• /
• 1994.04a
• /
• pp.157-157
• /
• 1994

원자력발전소 운영의 중추적 역할을 담당하고 있는 운전원과 발전소시스템 사이에서 발생하는 인간공학적 요인(인적요인)은 다중방호벽의 존재와 자동화 기술의 확대에도 불구하고 원전의 가동 성 및 안전성을 위협하는 최대의 요인이다. 최근 원자력발전소 시스템에 고도화된 전자공학 및 인공 지능기술 등이 반영되고 있는 추세이나 이러한 기술의 도입이 운전원과의 복합적 상호작용관점에서 원전의 안전성과 효율성에 적합한가를 실험적으로 평가할 수 있는 실험평가기술의 확보가 필요한 실정이다. 한국원자력연구소에서는 차세대 주제어실의 설계 및 평가를 위한 실험적 자료의 생성 및 설계 대안의 평가를 위한 기술확보라는 목적을 가지고 1992년도부터 수행하고 있다. 1992년도(1차년 도)에는 새로운 주제어실에서 실험적으로 평가해야 할 평가항목을 구체화하였고, 4년간의 연구추진 내용을 설정하였다. 기존의 원자력산업계에서 요구하고 있는 인가/허가 요건, 사업자요건서, 인간 공학분야에서 제기하고 있는 문제점 등을 분석하여 10개의 실험평가항목을 도출하였으며, 실험평가 항목을 실제로 실험을 통하여 연구하기 위한 장비 및 설비 그리고 소요기술 등을 고려하여 연구방향을 설정하였다. 1993년도(2차년도)에는 차세대 주제어실의 특징을 규명하고 실험연구의 대상시스템을 설정하였다. 설정된 대상시스템을 기능별로 분석하여 설계변수를 도출하였으며, 인간공학 실험실의 구축에 필수적인 원자력발전소 시뮬레이터의 기능요건 및 실험실의 구성요건 등을 개발하고 있다. 3차년도부터는 인간공학실험을 수행하면서 자료분석체계의 개발, 원전직무 시나리오의 개발, 측정방법의 개발, 인간공학 실험실의 설계, 구축 및 검증, 평가기법 연구, 실시간 자료수집체계의 개발 등을 수행할 예정이며, 연구종료시점인 1996년도(5차년도)에는 원자력발전소 주제어실의 인간공학적 평가를 위한 실험 환경의 구축 및 실험평가기술의 확립이라는 목표가 달성된다.하는 것으로 간주된다. 2. KR 53234 10mg/kg 정맥투여후의 최고혈중농도는 1.14ug/ml, 반감기는 0.50hr, 분포용적은 2.2L이었다. 20mg/kg 경구 투여시의 최소 혈중 농도는 0.33 ug/ml, 소실반감기는 1.5시간, AUC는 0.942ug.hr/ml, 분포용적 11L, Ka는 3.05 $hr^{－1}$ 그리고 Cl는 5.3L/hr/kg이었다. 이는 KR 53170에서와 같이 매우 적은량이 흡수되고 배설되었다. 3. KR 53170의 혈청단백 결합율은 5-500 ug/ml 범위에서 78.7-86.2%이었고 KR 53234의 혈청단백결합율은 5-100 ug/ml 범위에서 79.6-71.2%이었다.이었다.tic techniques, and we have recently cloned two of the major subunits; some of the data will be presented.LIFO, 우선 순위 방식등을 선택할 수 있도록 확장하였다. SIMPLE는 자료구조 및 프로그램이 공개되어 있으므로 프로그래머가 원하는 기능을 쉽게 추가할 수 있는 장점도 있다. 아울러 SMPLE에서 새로이 추가된 자료구조와 함수 및 설비제어 방식등을 활용하여 실제 중형급 시스템에 대한 시뮬레이션 구현과 시스템 분석의 예를 보인다._3$", chain segment, with the activation energy of carriers from the shallow trap with 0.4[eV], in he amorphous regions.의 증발산율은 우기의 기상자료를 이용하여 구한 결과 0.05 - 0.10 mm/hr 의 범위로서 이로 인한 강우손실량은 큰 의미가 없음을 알았다.재발이 나타난 3례의 환자를 제외한 9례 (75%)에서는 현재까지 재발소견을 보이지 않고 있다. 이러한 결과는 다른 보고자들과 유사한 결과를 보이고 있

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.41 no.2
• /
• pp.35-45
• /
• 2023

First, The spatial distribution characteristics of old big trees were analyzed using ArcGIS figures by combining basic information such as species and ages of old big trees in Handan City, which were compiled by the local bureau of landscaping. The types of species, distribution by ages of trees, ownership status, growth status, and diversity status were comprehensively analyzed. Statistically, Styphnolobium, Acacia, Gleditsia, and Albizia of Fabaceae accounted for the majority, of which Sophora japonica accounted for the highest proportion. Sophora japonica is widely and intensively distributed to each prefecture and district in Handan city. According to the age and distribution, the old big trees over 1000 years old were mainly Sophora japonica, Zelkova serrata, Juniperus chinensis, Morus australis Koidz., Dalbergia hupeana Hance, Ceratonia siliqua L., and Pistacia chinensis, and Platycladus orientalis. Second, as found in each type of old big tree status, various types of old big tree status were investigated, the protection management system, protection management process, and protection management benefits were studied, and the protection of old big tree was closely related to the growth environment. Currently, the main driving force behind the protection of old big trees is the worship of old big trees. By depositing its sacredness to the old big tree and sublimating the natural character that nature gave to the old big tree into a guiding consciousness of social activities, nature's "beauty" and personality's "goodness" are well combined. The protection state of the old big tree is closely related to the degree of interaction with the surrounding environment and the participation of various cultures and subjects. In the process of continuously interacting with the surrounding environment during the long-term growth of old big trees, it seems that a natural sanctuary was formed around old big trees in the process of voluntarily establishing a "natural-cultural-scape" system involving bottom-up and top-down cross-regions, multicultural and multi-subjects. Third, China focused on protecting and recovering old big trees, but the protection management system is poor due to a lack of comprehensive consideration of historical and cultural values, plant diversity significance, and social values of old big trees in the management process. Three indicators of space's regional characteristics, property and protection characteristics, and value characteristics can be found in the evaluation of the natural characteristics of old giant trees, which are highly valuable in terms of traditional consciousness management, resource protection practice, faith system construction, and realization of life community values. A systematic management system should be supported as to whether they can be protected and developed for a long time. Fourth, as the perception of protected areas is not yet mature in China, "natural sanctuary" should be treated as an important research content in the process of establishing a nature reserve system. The form of natural sanctuary management, which focuses on bottom-up community participation, is a strong supplement to the current type of top-down nature reserve management in China. Based on this, the protection of old giant trees should be included in the form of a nature reserve called a natural monument in the nature reserve system. In addition, residents of the area around the nature reserve should be one of the main agents of biodiversity conservation.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_3
• /
• pp.1827-1836
• /
• 2022

Disasters that occur unexpectedly are difficult to predict. In addition, the scale and damage are increasing compared to the past. Sometimes one disaster can develop into another disaster. Among the four stages of disaster management, search and rescue are carried out in the response stage when an emergency occurs. Therefore, personnel such as firefighters who are put into the scene are put in at a lot of risk. In this respect, in the initial response process at the disaster site, robots are a technology with high potential to reduce damage to human life and property. In addition, Light Detection And Ranging (LiDAR) can acquire a relatively wide range of 3D information using a laser. Due to its high accuracy and precision, it is a very useful sensor when considering the characteristics of a disaster site. Therefore, in this study, development and experiments were conducted so that the robot could perform real-time monitoring at the disaster site. Multi-sensor module was developed by combining LiDAR, Inertial Measurement Unit (IMU) sensor, and computing board. Then, this module was mounted on the robot, and a customized Simultaneous Localization and Mapping (SLAM) algorithm was developed. A method for stably mounting a multi-sensor module to a robot to maintain optimal accuracy at disaster sites was studied. And to check the performance of the module, SLAM was tested inside the disaster building, and various SLAM algorithms and distance comparisons were performed. As a result, PackSLAM developed in this study showed lower error compared to other algorithms, showing the possibility of application in disaster sites. In the future, in order to further enhance usability at disaster sites, various experiments will be conducted by establishing a rough terrain environment with many obstacles.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.6
• /
• pp.65-73
• /
• 2016

Peer-to-Peer (P2P) systems with super-peer overlay networks show combined advantages of both hybrid and pure P2P systems. Super-peer is a special peer acting as a server to a cluster of generic peers. Organizing a super-peer network is one of important issues for P2P systems with super-peer networks. Conventional P2P systems are based on two-level hierarchies of peers. One is a layer for generic peers and the other is for super-peers. And it is usual that super-peer networks have forms of random graphs. However, for accommodating a large-scale collection of generic peers, the super-peer network has also to be extended. In this paper, we propose a scheme of hierarchically constructing super-peer networks for large-scale P2P systems. At first, a two-level tree, called a simple super-peer network, is proposed, and then a scheme of generalizing and then extending the simple super-peer network to multi-level super-peer network is presented to construct a large-scale super-peer network. We call it an extended super-peer network. The simple super-peer network has several good features, but due to the fixed number of levels, it may have a scalability problem. Thus, it is extended to k-level tree of a super-peer network, called extended super-peer network. It shows good scalability and easy management of generic peers for large scale P2P system.