• Proceedings of the Korean Society of Computer Information Conference
• /
• 2015.07a
• /
• pp.250-251
• /
• 2015

본 논문에서는 자기 조직화 기능을 갖는 Kohonen의 SOM(Self organization map) 신경회로망과 주어지는 데이터에 따라 초기의 클러스터 개수를 설정하여 처리하는 수정된 K-Means 알고리즘을 결합한 Hybrid Kohonen Network 를 제안한다. 또한, 실제의 항공영상에 적용하여 고전적인 K-Means 알고리즘 및 고전적인 SOM 알고리즘보다 우수함을 보인다.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.6
• /
• pp.315-320
• /
• 2017

Estimating the correct pose of augmented objects on the real camera view efficiently is one of the most important questions in image tracking area. In computer vision, Homography is used for camera pose estimation in augmented reality system with markerless. To estimating Homography, several algorithm like SURF features which extracted from images are used. Based on extracted features, Homography is estimated. For this purpose, RANSAC algorithm is well used to estimate homography and DCS-RANSAC algorithm is researched which apply constraints dynamically based on Constraint Satisfaction Problem to improve performance. In DCS-RANSAC, however, the dataset is based on pattern of feature distribution of images manually, so this algorithm cannot classify the input image, pattern of feature distribution is not recognized in DCS-RANSAC algorithm, which lead to reduce it's performance. To improve this problem, we suggest the KCS-RANSAC algorithm using K-means clustering in CS-RANSAC to cluster the images automatically based on pattern of feature distribution and apply constraints to each image groups. The suggested algorithm cluster the images automatically and apply the constraints to each clustered image groups. The experiment result shows that our KCS-RANSAC algorithm outperformed the DCS-RANSAC algorithm in terms of speed, accuracy, and inlier rate.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.46-53
• /
• 2011

Nonlocal means denoising algorithm is one of the most widely used denoising algorithm. Because it performs well, and the theoretic idea is intuitive and simple. However the conventional nonlocal means algorithm has still some problems such as noise remaining in the denoised flat region and blurring artifacts in the denoised edge and pattern region. Thus many improved algorithms based on nonlocal means have been proposed. In this paper, we proposed new improved nonlocal means denoising algorithm by weight update through weights sorting and newly defined threshold. Updated weights can make weights more refined and definite, and denoising is possible without that artifacts. Experimental results including comparisons with conventional algorithms for various noise levels and test images show the proposed algorithm has a good performance in both visual and quantitative criteria.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.259-262
• /
• 2002

벡터양자화에서 주어진 학습벡터를 가장 잘 대표할 수 있는 코드벡터의 집합인 코드북을 구하는 것은 가장 중요한 문제이다. 이러한 코드북을 구하는 알고리즘 중에서 가장 대표적인 방법은 K-means 알고리즘으로 그 성능이 초기 코드북에 크게 의존한다는 문제점을 가지고 있어 여러 가지 초기 코드북을 설계하는 알고리즘이 제안되어 왔다. 본 논문에서는 splitting 방법을 이용한 수정된 초기 코드북 생성 알고리즘을 제안하고자 한다. 제안된 방법에서는 기존외 splitting 방법을 적용하여 초기 코드북을 생성하되, 미소분리 과정 시 학습벡터의 수렴 빈도가 가장 낮은 코드벡터를 제거하고 수렴 빈도가 가장 높은 코드벡터를 미소분리 하여 수렴 빈도가 가장 낮은 코드벡터와 대체해가며 초기 코드북을 설계 한다. 제안된 방법의 적용온 기존 방법에서 MSE(mean square error)의 감소율이 가장 작은 미소분리 과정에서 시작하여 원하는 코드북 크기를 얻을 때까지 반복한다. 제안된 방법으로 생성된 초기 코드북을 사용하여 K-means 알고리즘을 수행한 결과 기존의 splitting 방법으로 생성된 초기 코드북을 사용한 경우보다 코드북의 성능이 향상되었다.

• Proceedings of the IEEK Conference
• /
• 2003.07d
• /
• pp.1557-1560
• /
• 2003

본 논문에서는 이 문서 클러스터링 방법 중 계층적 방법인 Kmeans 클러스터링 알고리즘을 이용하여 문서를 클러스터링 하고자 한다. 기존의 Kmeans 클러스터링 알고리즘은 문서의 수가 많을 경우 하나의 클러스터링에 너무 많은 문서들이 할당되는 문제점이 있다. 이 치우침을 완화하고자 각 클러스터링에 할당된 문서 수에 따라서 문서에 가중치를 부여한 후 다시 클러스터링을 하는 방법을 제안하였다. 실험 결과는 정확률, 재현율을 결합한 조화 평균(F-measure)을 사용하여 평가하였으며 기존 알고리즘보다 9％이상의 성능 향상을 나타냈다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.4
• /
• pp.416-426
• /
• 2003

In this paper we propose a new method for choosing the initial values of Expectation-Maximization(EM) algorithm that has been used in various applications for clustering. Conventionally, the initial values were chosen randomly, which sometimes yields undesired local convergence. Later, K-means clustering method was employed to choose better initial values, which is currently widely used. However the method using K-means still has the same problem of converging to local points. In order to resolve this problem, a new method of initializing values for the EM process. The proposed method not only strengthens the characteristics of EM such that the number of iteration is reduced in great amount but also removes the possibility of falling into local convergence.

• Journal of KIISE:Software and Applications
• /
• v.30 no.11
• /
• pp.1072-1079
• /
• 2003

This study is purported for suggesting a new clustering algorithm that enables incremental categorization of numerous documents. The suggested algorithm adopts the natures of the spherical k-means algorithm, which clusters a mass amount of high-dimensional documents, and the fuzzy ART(adaptive resonance theory) neural network, which performs clustering incrementally. In short, the suggested algorithm is a combination of the spherical k-means vector space model and concept vector and fuzzy ART vigilance parameter. The new algorithm not only supports incremental clustering and automatically sets the appropriate number of clusters, but also solves the current problems of overfitting caused by outlier and noise. Additionally, concerning the objective function value, which measures the cluster's coherence that is used to evaluate the quality of produced clusters, tests on the CLASSIC3 data set showed that the newly suggested algorithm works better than the spherical k-means by 8.04% in average.

• The KIPS Transactions:PartD
• /
• v.18D no.5
• /
• pp.339-342
• /
• 2011

Recently, researches are studied in developing efficient techniques for accessing, querying, and storing XML documents which are frequently used in the Internet. In this paper, we propose a new method to cluster XML documents efficiently. We use a K-means algorithm with a Principal Component Analysis(PCA) to cluster XML documents after they are represented by vectors in the feature vector space by transferring them as names and levels of the elements of the corresponding trees. The experiment shows that our proposed method has a good result.

• Journal of KIISE
• /
• v.45 no.3
• /
• pp.288-293
• /
• 2018

There have recently been several studies investigating how to apply a privacy preserving technique to publish data. Differential privacy can protect personal information regardless of an attacker's background knowledge by adding probabilistic noise to the original data. To perform differentially private k-means clustering, the existing algorithm builds a differentially private histogram and performs the k-means clustering. Since it constructs an equi-width histogram without considering the distribution of data, there are many buckets to which noise should be added. We propose a k-means clustering algorithm using a quad-tree that captures the distribution of data by using a small number of buckets. Our experiments show that the proposed algorithm shows better performance than the existing algorithm.

• Journal of IKEEE
• /
• v.22 no.3
• /
• pp.658-663
• /
• 2018

In this paper, we consider node deployment algorithms for the sequence-based localization (SBL) which is recently employed for in-door positioning systems, Whereas previous node selection or deployment algorithms seek to place nodes at centrold of the region where more targets are likely to be found, we observe that the boundaries dividing such regions can be good locations for the nodes in SBL systems. Motivated by this observation, we propose an efficient node deployment algorithm that determines the boundaries by using the well-known K-means algorithm and find the potential node locations based on the bi-section method for low-complexity design. We demonstrate through experiments that the proposed algorithm achieves significant localization performance over random node allocation with a substantially reduced complexity as compared with a full search.