• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.4
• /
• pp.267-273
• /
• 2011

In this paper, we proposed the method to detect brain tumor region in MR images. Our method is composed of 3 parts, detection of tumor slice, detection of tumor region and tumor boundary detection. In the tumor slice detection step, a slice which contains tumor regions is distinguished using symmetric analysis in 3D brain volume. The tumor region detection step is the process to segment the tumor region in the slice distinguished as a tumor slice. And tumor region is finally detected, using spatial feature and symmetric analysis based on the cluster information. The process for detecting tumor slice and tumor region have advantages which are robust for noise and requires less computational time, using the knowledge of the brain tumor and cluster-based on symmetric analysis. And we use the level set method with fast marching algorithm to detect the tumor boundary. It is performed to find the tumor boundary for all other slices using the initial seeds derived from the previous or later slice until the tumor region is vanished. It requires less computational time because every procedure is not performed for all slices.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.335-339
• /
• 2006

자궁 경부 세포진 영상의 핵 영역 분할은 자궁 경부암 자동화 검색 시스템의 가장 어렵고도 중요한 분야로 알려져 있다. 자궁 경부 세포진 영상은 배경과 세포의 영역이 확실히 구분되지 않는 경우가 많기 때문에 이들을 확실히 구분하는 것이 매우 중요하다. 본 논문에서는 이러한 문제점을 해결하기 위해 자궁 경부 세포진 영상에서 Region growing 기법을 적용하여 세포 영상을 분할한다. Region growing 기법은 화소간의 유사도를 측정하여 영역을 확장하여 분할하는 방법이다. 세포와 배경이 분할된 영상을 일정 임계값을 이용하여 영상을 이진화 한 후, 8방향 윤곽선 추적 알고리즘을 이용해 세포 영역을 추출한다. 추출된 세포 영역을 원 영상인 RGB 컬러로 변환한 후에 K-means 알고리즘을 적용하여 각 세포 영역의 RGB 화소를 R, G, B 채널로 각각 분리하여 클러스터링한다. 클러스터링된 각각의 R, G, B 채널의 클러스터 값을 이용하여 HSI 모델로 변환시킨 후에 세포핵 영역의 Hue 정보를 추출한다. 추출된 세포핵의 특징을 오류 역전파 알고리즘을 적용하여 정상 세포와 비정상 세포를 분류하고 인식한다.

• The Journal of the Korea Contents Association
• /
• v.14 no.10
• /
• pp.20-31
• /
• 2014

Recently, cache sharing methods have been studied in order to effectively reply to user requests in mobile P2P networks. In this paper, we propose a cache sharing scheme based on a cluster considering the peer connectivity in mobile P2P networks. The proposed scheme shares caches by making a cluster that consists of peers preserving the connectivity among them for a long time. The proposed scheme reduces data duplication to efficiently use the cache space in a cluster. The cache space is divided into two parts with a data cache and a temporary cache for a cache space. It is possible to reduce the delay time when the cluster topology is changed or the cache data is replaced utilizing a temporary cache. The proposed scheme checks the caches of peers in a route to a cluster header and the caches of one-hop peers in order to reduce the communication cost. It is shown through performance evaluation that the proposed scheme outperforms the existing schemes.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.30 no.6
• /
• pp.21-30
• /
• 2002

A preconditioned Navier-Stokes code was parallelized by the domain decomposition technique, and the accuracy of the parallelized code was verified through a comparison with the result of a sequential code and experimental data. Parallel performance of the code was examined on a Myrinet based PC-cluster and a Fast-Ethernet system. Speed-up ratio was examined as a major performance parameter depending on the number of processor and the network communication topology. In this test, Myrinet system shows a superior parallel performance to the Fast-Ethernet system as was expected. A test for the dependency on problem size also shows that network communication speed in a crucial factor for parallel performance, and the Myrinet based PC-cluster is a plausible candidate for high performance parallel computing system.

• Journal of Digital Convergence
• /
• v.14 no.1
• /
• pp.203-209
• /
• 2016

In this paper, we introduce a fuzzy inference systems for nonlinear inference using fuzzy cluster. Typically, the generation of fuzzy rules for nonlinear inference causes the problem that the number of fuzzy rules increases exponentially if the input vectors increase. To handle this problem, the fuzzy rules of fuzzy model are designed by dividing the input vector space in the scatter form using fuzzy clustering algorithm which expresses fuzzy cluster. From this method, complex nonlinear process can be modeled. The premise part of the fuzzy rules is determined by means of FCM clustering algorithm with fuzzy clusters. The consequence part of the fuzzy rules have four kinds of polynomial functions and the coefficient parameters of each rule are estimated by using the standard least-squares method. And we use the data widely used in nonlinear process for the performance and the nonlinear characteristics of the nonlinear process. Experimental results show that the non-linear inference is possible.

• Journal of KIISE:Computer Systems and Theory
• /
• v.28 no.11
• /
• pp.549-560
• /
• 2001

Cluster system provides attractive scalability in terms of compution power and memory size. With the advances in high speed computer network technology, cluster systems are becoming increasingly competitive compared to expensive MPPs (massively parallel processors). Load balancing is very important issue since an inappropriate scheduling of tasks cannot exploit the true potential of the system and can offset the gain from parallelization. In parallel processing program, it is difficult to predict the load of each task before running the program. Furthermore, tasks are interdependent each other in many ways. The dynamic load balancing algorithm, which evaluates each processor's load in runtime, partitions each task into the appropriate granularity and assigns them to processors in proportion to their performance in cluster systems. However, if the communication cost between processing nodes is expensive, it is not efficient for all nodes to attend load balancing process. In this paper, we restrict a processor that attend load balancing by the communication cost and the deviation of its load from the average. We simulate various models of the cluster system with parameters such as communication cost, node number, and range of workload value to compare existing load balancing methods with the proposed dynamic algorithms.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.284-284
• /
• 2023

도시 침수의 발생과 규모는 도시 유역이 가지는 짧은 도달 시간으로 인하여 주로 시단위 이하의 짧은 지속시간의 강우의 극한 및 변동성에 따라 결정된다. 미래 기간에 대하여 도시 수문 시스템의 적정성을 평가하기 위해서는 기후변화에 따른 시단위 이하 강우의 특성을 살펴보아야한다. 그러나 기후변화 영향 평가 도구로 활용되는 기후 모형들은 대부분 일단위의 결과물을 제공하여 시단위 이하의 미세 규모 강우의 특성을 나타낼 수 없다. 이에 따라 본 연구에서는 기후 모형 모의 결과와 포아송 클러스터 강우 모형을 이용하여 미래 시단위 이하 강우 시계열을 모의하는 방법을 제안한다. 첫째로, 포아송 클러스터 기반 강우 생성 알고리즘과 폭풍우 재배열 알고리즘을 결합한 최신 모형을 선정하였다. 해당 모형은 광범위한 시간 규모에서 관측된 강우량의 주요 통계와 극값을 재현할 수 있는 모형이다. 그 다음 강우 모형에 적합시킬 관측 강우량 통계(평균, 분산, 공분산, 왜도, 우기 비율)를 계산하였다. 둘째, 강우 통계 간의 선형 관계를 도출하였다. 여기서는 클러스터에 있는 모든 관측소의 통계를 사용하여 회귀의 신뢰도를 높였다. 셋째, 강우 평균 조정을 위한 Change Factor는 제어(2000~2019년) 및 미래(2041~2070년) 기간의 기후 모형 자료를 사용하여 계산하였다. 넷째, 조정된 15분 강우 평균은 관측 평균에 Change Factor을 곱하여 계산하고 조정된 강우 평균과 통계 간의 관계를 사용하여 미래 강우 통계 세트를 추정하였다. 여러 통계 세트를 생성한 후 마지막으로 미래 통계에 강우 모형을 적합시켜 최종적으로 미래 시단위 이하 강우 시계열을 모의하였다. 이 방법은 CMIP6에 참여하는 기후 모델의 기후 예측 데이터를 사용하여 용산(415) 및 동래(940) AWS 관측소에 적용되었다. 두 관측소의 미래 강우 모의 결과, 시단위 이하 시간 규모에서 극값이 증가하는 추세를 보였다.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.1
• /
• pp.22-33
• /
• 1999

In this paper, we propose a new multi-way circuit partitioning system that partition large circuits to progrmmable circuit board which consist of FPGAs and interconnect components. Here the routing topology among the chips are predetermined and the number of available interconnections are fixed. Since the given constraints are difficult to be satisfied by the previous partition method, we suggest a new multi-way partition method by target restriction that considers all the constraints for the given board. To speed up, we construct a multi-level cluster tree for hierarchical partitioning. Experimental results for several benchmarks show that the our partition method partition them by satisfying all the given constraints and it used up to 10 % fewer interconnections among the chips than the previous K-way partition method.

• The KIPS Transactions:PartA
• /
• v.12A no.7 s.97
• /
• pp.613-620
• /
• 2005

To provide reliable differentiated service for internet clients, many studies on differentiated service are widely done. According to the types of the request of the clients in cluster-based web systems, SLA (Service Level Agreement) between service provider and the clients is considered. In this Paper, we study primary-backup web switch in the web systems that distribute the request of clients to each of web servers and also study performance isolation and admission control schemes of heterogeneous web systems which distinguish the servers into each different function domains. We analyze the availability, the response time and the performance of admission control through various simulated experiments on the proposed web systems.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.6
• /
• pp.493-497
• /
• 2016

Infrared (IR) imaging is capable of detecting targets that are not visible at night, thus it has been widely used for the security and defense system. However, the quality of the IR image is often degraded by low resolution and noise corruption. This paper addresses target segmentation with the IR image. Multiple regions of interest (ROI) are extracted by the multi-level segmentation and targets are segmented from the individual ROI. Each level of the multi-level segmentation is composed of a k-means clustering algorithm an expectation-maximization (EM) algorithm, and a decision process. The k-means clustering algorithm initializes the parameters of the Gaussian mixture model (GMM) and the EM algorithm iteratively estimates those parameters. Each pixel is assigned to one of clusters during the decision. This paper proposes the selection and the merging of the extracted ROIs. ROI regions are selectively merged in order to include the overlapped ROI windows. In the experiments, the proposed method is tested on an IR image capturing two pedestrians at night. The performance is compared with conventional methods showing that the proposed method outperforms others.