• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 춘계 학술대회 학술발표 논문집
• /
• pp.235-238
• /
• 2003

Fuzzy C-Means(FCM) 알고리즘은 초기 군집 중심의 개수와 위치에 따라 군집 결과의 성능차이가 많이 나타난다. 하지만 일반적인 경우에 군집 중심의 개수는 분석가의 주관에 의해 결정되고, 임의적으로 결정되기 때문에 원래 데이터의 구조와는 무관하게 수행되어 최적화된 군집화 수행을 실행하지 못하는 경우가 발생하게 된다. 따라서 본 논문에서는 원래의 데이터의 구조에 좀더 근접한 퍼지 군집화를 수행하기 위하여 격자를 바탕으로 한 데이터의 밀도를 이용한 FCM을 제안하고, 이러한 밀도 기반 FCM에 의해 결정된 군집의 합병 기법을 제안하였다. N-차원의 데이터 공간을 N-차원의 격자로 나누고, 초기 군집 중심의 개수와 위치는 각 격자의 밀도를 바탕으로 결정된다. 초기화 이후에 각 격자 내부에서 FCM을 이용하여 군집화를 수행하고, 계속해서 이웃 격자의 군집결과에 대하여 군집간의 유사도 측도를 이용하여 군집 합병을 수행함으로써 데이터의 자연적인 구조에 근접한 군집화를 수행하였다. 제안된 군집화 합병 기법의 향상된 성능은 UCI Machine Learning Repository 데이터를 이용하여 확인하였다.

• 대한산업공학회지
• /
• 제29권1호
• /
• pp.90-99
• /
• 2003

Object clustering, which makes classification for a set of objects into a number of groups such that objects included in a group have similar characteristic and objects in different groups have dissimilar characteristic each other, has been exploited in diverse area such as information retrieval, data mining, group technology, etc. In this study, an object-clustering problem with similarity coefficients between objects is considered. At first, an evaluation function for the optimization problem is defined. Then, a genetic algorithm and local optimization technique based on heuristic method are proposed and used in order to obtain near optimal solutions. Solutions from the genetic algorithm are improved by local optimization techniques based on object relocation and cluster merging. Throughout extensive experiments, the validity and effectiveness of the proposed algorithms are tested.

• Journal of information and communication convergence engineering
• /
• 제5권1호
• /
• pp.59-63
• /
• 2007

In this paper, a CLB-based CPLD low-power technology mapping algorithm considered a Trade-off is proposed. To perform low-power technology mapping for CPLDs, a given Boolean network has to be represented in a DAG. The proposed algorithm consists of three steps. In the first step, TD(Transition Density) calculation has to be performed. Total power consumption is obtained by calculating the switching activity of each node in a DAG. In the second step, the feasible clusters are generated by considering the following conditions: the number of inputs and outputs, the number of OR terms for CLB within a CPLD. The common node cluster merging method, the node separation method, and the node duplication method are used to produce the feasible clusters. In the final step, low-power technology mapping based on the CLBs packs the feasible clusters. The proposed algorithm is examined using SIS benchmarks. When the number of OR terms is five, the experiment results show that power consumption is reduced by 30.73% compared with TEMPLA, and by 17.11 % compared with PLA mapping.

• 융합신호처리학회논문지
• /
• 제16권1호
• /
• pp.9-14
• /
• 2015

Outlier detection and removal is a crucial step needed for various image processing applications such as image registration. Random Sample Consensus (RANSAC) is known to be the best algorithm so far for the outlier detection and removal. However RANSAC requires a cosiderable computation time. To drastically reduce the computation time while preserving the comparable quality, a outlier detection and removal method based on modified K-means is proposed. The original K-means was conducted first for matching point pairs and then cluster merging and member exclusion step are performed in the modification step. We applied the methods to various images with highly repetitive patterns under several geometric distortions and obtained successful results. We compared the proposed method with RANSAC and showed that the proposed method runs 3~10 times faster than RANSAC.

• 대한원격탐사학회지
• /
• 제20권3호
• /
• pp.215-225
• /
• 2004

A multistage hierarchical clustering technique, which is an unsupervised technique, was suggested in this paper for classifying large remotely-sensed imagery. The multistage algorithm consists of two stages. The 'local' segmentor of the first stage performs region-growing segmentation by employing the hierarchical clustering procedure of CN-chain with the restriction that pixels in a cluster must be spatially contiguous. The 'global' segmentor of the second stage, which has not spatial constraints for merging, clusters the segments resulting from the previous stage, using the conventional agglomerative approach. Using simulation data, the proposed method was compared with another hierarchical clustering technique based on 'mutual closest neighbor.' The experimental results show that the new approach proposed in this study considerably increases in computational efficiency for larger images with a low number of bands. The technique was then applied to classify the land-cover types using the remotely-sensed data acquired from the Korean peninsula.

• 천문학회보
• /
• 제44권1호
• /
• pp.38.1-38.1
• /
• 2019

Globular clusters (GCs) are one of the excellent tools to trace the assembly history of their host galaxies. Especially, the ages and abundances of the GCs give important clues about the star formation epochs and merging progenitors. We investigate the stellar population of the GCs in M87 based on a stacking analysis using about 900 MMT/Hectospec spectra of the GCs. We measure the ages, [Z/H], and [a/Fe] from the stacked spectra of the GCs within radial bins based on Lick indices. We find clear radial gradients for [Z/H] and [a/Fe] in the GC system. In addition to the radial trends, we investigate the stellar populations of the GC subgroups divided according to colors, radial velocities, and spatial locations. We discuss the formation history of M87 based on the stellar populations of the GCs.

• 대한전자공학회논문지SP
• /
• 제43권6호
• /
• pp.20-27
• /
• 2006

Multi-level thresholding은 영상 분할 방법 중 하나로 널리 이용되고 있지만 대부분의 기존 논문들은 응용 분야에 직접적으로 이용되기에는 적합하지 않거나 영상 분할 단계까지 확장되지 않고 있다. 본 논문에서는 영상 분할을 위한 multi-level thresholding 방안으로써 영역 단위의 multi-level thresholding을 제안한다. 먼저, 영상의 색상별 성분에 대해서 EWFCM(Entropy-based Weighted Fuzzy C-Means) 알고리즘을 적용하여 2개의 군집으로 분류한 후 코드 영상을 생성한다. EWFCM 알고리즘은 화소들에 대한 공간 정보를 추가한 개선된 FCM 알고리즘으로 영상 내 존재하는 잡음을 제거한다. 그리고 코드 영상에 존재하는 군집의 수를 감소함으로써 좀 더 나은 영상 분할 결과를 얻을 수 있으며 군집의 감소는 하나의 군집내에 존재하는 영역들과 나머지 군집들간의 유사도를 기반으로 영역을 재분류함으로써 처리된다. 그러나 영상에는 여전히 많은 영역들이 존재하기 때문에 이를 해결하기 위한 하나의 후처리 방안으로써 영역간의 Kullback-Leibler 거리값을 기반으로 Bayesian 알고리즘에 의한 영역 합병을 수행한다. 실험 결과 제안한 영역 기반의 multi-level thresholding은 기존 방법이나 화소나 군집 기반의 multi-level thresholding보다 좋은 분할 결과를 보였으며 Bayesian 알고리즘을 이용한 후처리 방안에 의해 좀 더 나은 결과를 보였다.

• 한국정보과학회논문지:데이타베이스
• /
• 제31권4호
• /
• pp.409-421
• /
• 2004

데이타베이스 공유 시스템(DSS)은 고성능 트랜잭션 처리를 위하여 여러 개의 처리 노드를 연결한 구조로서, 각 노드는 데이타베이스를 저장한 디스크를 공유한다. DSS를 구성하는 노드들이 고장날 경우 데이타베이스를 정확한 상태로 복구하기 위한 회복 과정이 필요한데 DSS에서 회복 작업은 하나의 노드로 구성된 일반적인 데이타베이스 시스템보다 많은 시간이 소요된다. 그 이유는 데이타베이스를 회복하기 위해 여러 노드에 나누어 저장된 로그들을 병합하여야 하며, 병합된 로그들을 이용하여 REDO 작업을 수행하여야 하기 때문이다. 본 논문에서는 Oracle 9i Real Application Cluster (ORAC)에서 제안된 캐쉬 연합 알고리즘의 성능을 개선한 2VC(Two Version Caching) 알고리즘을 제안한다. 2VC는 단일 노드 고장에 대한 회복 작업에서 로그 병합 과정을 생략할 수 있으므로 빠른 데이타베이스 회복을 지원할 수 있다는 장점을 갖는다. 뿐만 아니라, ORAC에서 발생하는 불필요한 디스크 기록 오버헤드를 줄임으로써 정상적인 트랜잭션 처리의 성능을 향상시킬 수 있다.

• 천문학회보
• /
• 제38권2호
• /
• pp.73.2-73.2
• /
• 2013

In the current ${\Lambda}CDM$ hierarchical merging paradigm, a galaxy like the Milky Way formed by numerous mergers of ancient subsystems. Most of the relics of these building blocks, however, are yet to be discovered or identified. Recent progress in the Milky Way globular cluster research is throwing new light on this perspective. The discoveries of multiple stellar populations having different heavy element abundances in some massive globular clusters are suggesting that they are most likely the remaining cores or relics of disrupted dwarf galaxies. In this talk, we will report our progress in the (1) narrow-band photometry, (2) low-resolution spectroscopy, and (3) population modeling for this growing group of peculiar globular clusters.

• 한국환경보건학회지
• /
• 제38권5호
• /
• pp.369-379
• /
• 2012

Objectives: Spatial analysis is useful for understanding complicated causal relationships. This paper focuses trends and appling methods for spatial analysis associated with environmental asbestos exposure. Methods: Literature review and reflection of experience of authors were conducted to know academic background of spatial analysis, appling methods on epidemiology and asbestos exposure. Results: Spatial analysis based on spatial autocorrelation provides a variety of methods through which to conduct mapping, cluster analysis, diffusion, interpolation, and identification. Cause of disease occurrence can be investigated through spatial analysis. Appropriate methods can be applied according to contagiousness and continuity. Spatial analysis for asbestos exposure source is needed to study asbestos related diseases. Although a great amount of research has used spatial analysis to study exposure assessment and distribution of disease occurrence, these studies tend to focus on the construction of a thematic map without different forms of analysis. Recently, spatial analysis has been advanced by merging with web tools, mobile computing, statistical packages, social network analysis, and big data. Conclusions: Because the trend in spatial analysis has evolved from simple marking into a variety of forms of analyses, environmental researchers including asbestos exposure study are required to be aware of recent trends.