• Journal of Korea Multimedia Society
• /
• v.3 no.4
• /
• pp.339-346
• /
• 2000

Medical image is analyzed to get an anatomical information for diagnostics. Segmentation must be preceded to recognize and determine the lesion more accurately. In this paper, we propose automatic segmentation algorithm for MR brain images using T1-weighted, T2-weighted and PD images complementarily. The proposed segmentation algorithm is first, extracts cerebrum images from 3 input images using cerebrum mask which is made from PD image. And next, find 3D clusters corresponded to cerebrum tissues using scale filtering and 3D clustering in 3D space which is consisted of T1, T2, and PD axis. Cerebrum images are segmented using FCM algorithm with its initial centroid as the 3D cluster's centroid. The proposed algorithm improved segmentation results using accurate cluster centroid as initial value of FCM algorithm and also can get better segmentation results using multi spectral analysis than single spectral analysis.

• Journal of IKEEE
• /
• v.23 no.2
• /
• pp.419-424
• /
• 2019

The FCM algorithm finds the optimal solution through iterative optimization technique. In particular, there is a difference in execution time depending on the initial center of clustering, the location of noise, the location and number of crowded densities. However, this method gradually updates the center point, and the center of the initial cluster is shifted to one side. In this paper, we propose a TI-FCM(Triangular Inequality-Fuzzy C-Means) clustering algorithm that determines the cluster center density by maximizing the distance between clusters using triangular inequality. The proposed method is an effective method to converge to real clusters compared to FCM even in large data sets. Experiments show that execution time is reduced compared to existing FCM.

• Journal of Biomedical Engineering Research
• /
• v.21 no.4
• /
• pp.333-338
• /
• 2000

In this paper, an automated segmentation algorithm is proposed for MR brain images using T1-weighted, T2-weighted, and PD images complementarily. The proposed segmentation algorithm is composed of 3 step. In the first step, cerebrum images are extracted by putting a cerebrum mask upon the three input images. In the second step, outstanding clusters that represent inner tissues of the cerebrum are chosen among 3-dimensional(3D) clusters. 3D clusters are determined by intersecting densely distributed parts of 2D histogram in the 3D space formed with three optimal scale images. Optimal scale image is made up of applying scale space filtering to each 2D histogram and searching graph structure. Optimal scale image best describes the shape of densely distributed parts of pixels in 2D histogram and searching graph structure. Optimal scale image best describes the shape of densely distributed parts of pixels in 2D histogram. In the final step, cerebrum images are segmented using FCM algorithm with its initial centroid value as the outstanding clusters centroid value. The proposed cluster's centroid accurately. And also can get better segmentation results from the proposed segmentation algorithm with multi spectral analysis than the method of single spectral analysis.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.9 no.5
• /
• pp.490-495
• /
• 1999

In this paper. a learning method VoD-EBP for neural networks is proposed, which learn patterns by error back propagation. Based on Voronoi diagram, the method initializes the weights of the neural networks systematically, wh~ch results in faster learning speed and alleviated local optimum problem. The method also shows better the reliability of the design of neural network because proper number of hidden nodes are determined from the analysis of Voronoi diagram. For testing the performance, this paper shows the results of solving the XOR problem and the parity problem. The results were showed faster learning speed than ordinary error back propagation algorithm. In solving the problem, local optimum problems have not been observed.

• Journal of the Korea Society for Simulation
• /
• v.17 no.3
• /
• pp.95-105
• /
• 2008

In this paper, we propose the DAMMC(Distributed Authentication Model using Multi-level Cluster) for wireless sensor networks. The proposed model is that one cluster header in m-layer has a role of CA(Certificate Authority) but it just authenticates sensor nodes in lower layer for providing an efficient authentication without authenticating overhead among clusters. In here, the m-layer for authentication can be properly predefined by user in consideration of various network environments. And also, the DAMMC uses certificates based on the threshold cryptography scheme for more reliable configuration of WSN. Experimental results show that the cost of generation and reconfiguration certification are decreased but the security performance are increased compared to the existing method.

• Annual Conference of KIPS
• /
• 2000.10b
• /
• pp.1433-1436
• /
• 2000

클러스터링에 있어서 k-means[7], DBSCAN[2], CURE[4], ROCK[5], PAM[8], 같은 기존의 알고리즘은 원형이나 타원형 등의 어느 고정된 모양에 의해 클러스터를 결정한다. 만약 클러스터 하려는 데이터의 분포가 우연히 알고리즘의 결정된 모양과 일치하면 정확한 해를 얻을 수 있다. 하지만 자연적인 데이터의 분포에서는 발생하기 어렵다. 데이터의 형태를 추적하여 이러한 문제점을 해결한 CHAMELEON[1] 알고리즘이 최근에 발표되었다. 하지만 모양에는 독립적이나 데이터의 양이 증가함에 따라 소요되는 시간이 폭발적으로 증가한다. 이것은 기존의 마이닝 데이터들이 대용량이라는 것을 고려하면 현실에 적용하기 힘든 문제점이 있다. 이러한 문제점을 해결하기 위해 본 논문에서는 K-means[7]]를 이용한 대표를 선출하는 방법으로 CHAMELEON[1]의 문제점 개선(EF-CHAMELEON)을 시도하였으며 여러 자연적인 형태의 도형들은 아주 작은 원형들의 집합으로 구성 될 수 있다는 생각을 기본으로 잡음에 영향을 받지 않을 정도로 아주 작은 초기 다수의 소형 클러스터를 K-mean을 이용하여 구성하고 이를 다시 크러스터간의 상대적인 거리를 이용하여 다시 머지 하는 방법으로 모양에 의존적인 문제를 해결하며 비교사 학습(unsupervised learning)에 충실하기 위해 임계값을 적용 적정 단계에서 알고리즘을 멈추게 한 ADF 알고리즘을 소개한다. 실험 데이터는 기존의 여러 클러스터링 알고리즘이 판별 할 수 없었던 다양한 모양을 가지고있는 2차원 배열을 사용하여 ADF. CHAMELEON[1], EF-CHAMELEON,의 성능을 비교하였다.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.2
• /
• pp.150-156
• /
• 2008

Many papers have shown that the hierarchical clustering method takes good-performance, but is limited because of its quadratic time complexity. In contrast, with a large number of variables, K-means reduces a time complexity. Think of the factor of simplify, high-quality and high-efficiency, we combine the two approaches providing a new system named CONDOR system with hierarchical structure based on document clustering using K-means algorithm. Evaluated the performance on different hierarchy depth and initial uncertain centroid number based on variational relative document amount correspond to given queries. Comparing with regular method that the initial centroids have been established in advance, our method performance has been improved a lot.

• Journal of Intelligence and Information Systems
• /
• v.15 no.1
• /
• pp.1-14
• /
• 2009

The Capacitated Vehicle Routing Problem (CVRP) is the problem that the vehicles stationed at central depot are to be optimally routed to supply customers with demands, satisfying vehicle capacity constraints. The CVRP is the NP-hard as it is a natural generalization of the Traveling Salesman Problem (TSP). In this article, we propose the heuristic algorithm, called the bisection seed detection method, to solve the CVRP. The algorithm is composed of 3-phases. In the first phase, we work out the initial cluster using the improved sweep algorithm. In the next phase, we choose a seed node in each initial cluster by using the bisection seed detection method, and we compose the rout with the nearest node from each seed. At this phase, we compute the regret value to decide the list of priorities for the node assignment. In the final phase, we improve the route result by using the tabu search and exchange algorithm. We compared our heuristic with different heuristics such as the Clark-Wright heuristic and the genetic algorithm. The result of proposed heuristic show that our algorithm can get the nearest optimal value within the shortest execution time comparatively.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.06a
• /
• pp.188-194
• /
• 2007

We propose a new ad hoc multicast routing protocol for based on the ontology scheme called inference network. Ontology knowledge-based is one of the structure of context-aware. We will have developed an algorithm that will design multi-hierarchy Layered networks to simulate a desired system.

• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2008.11a
• /
• pp.37-37
• /
• 2008

고온에서 안정적인 DLC막을 성막하기 위해 PECVD공정에서 실리콘을 첨가하여 제조하였다. 기존의 실리콘첨가 DLC막과는 다르게 고온에서 생성됨으로 마이크로 클러스터 형태의 DLC구조로서 disordered 영역이 넓게 존재하고 있어 I(D)/I(G)비에서의 변화가 있는 것이 관찰되었다. 실리콘 양이 증가할수록 값이 낮아지는 것이 관찰되는데 이는 실리콘량이 증가하면서 수소의 위치에 실리콘이 결합하면서 sp3 단일구조형태의 코팅 막을 만드는 것이 관찰된다. 고온 어닐링효과로 내부구조에서 다량의 sp2구조가 관찰되는 것으로서 DLC막이 어느 정도 흑연화되지만, 실리콘이 SiC에서 SiOx로 $SiO_2$와 SiOH막으로 바뀌는 면서 마찰계수가 낮은 DLC막을 유지할 것으로 기대되며, XPS와 FT-IR분석에 의해 이러한 상들의 존재를 관찰할 수 있었다. 특히 공정상 TMS이 증가하면 첨가된 Si에 의해 형성되는 막이 초기부터 OH기를 다량 포함하고 있는 것을 알 수 있었고, 온도 상승에 의해서 실리콘표층에 더욱 많은 SiOx계열의 물질이 생성되는 것이 명확하게 발견되었다.