• 한국정보과학회논문지:데이타베이스
• /
• 제28권3호
• /
• pp.344-356
• /
• 2001

본 논문에서는 시퀀스 데이터베이스에서 시간왜곡 변환(time warping)을 지원하는 서브시퀀스 탐색 문제를 다룬다. 서브시퀀스 탐색은 데이터 시퀀스의 평균 길이의 이차 함수로 성능이 저하된다. 이러한 문제를 해결하기 위하여 본 논문에서는 세그먼트 기반 서브시퀀스 탐색 기법(Segment-Based Approach for Subsequence Searches : SBASS)을 제안한다. SBASS는 데이터와 질의 시퀀스를 연속된 세그먼트들로 분할하여 다음의 두가지 조건을 만족하는 모든 데이터 시퀀스를 검색한다. (1) 세그먼트의 개수가 질의 시퀀스의 세그먼트 개수와 같다. (2) 모든 세그먼트 쌍 간의 거리가 주어진 오차 한도 이내이다. 제안된 세그먼트 분할 기법에서는 세그먼트가 서로 다른 길이를 갖도록 허용하며, 세그먼트 쌀간의 유사성의 척도로서 시간왜곡 변환 거리를 이용한다. 효율적인 유사 서브시퀀스 탐색을 위하여, 각 데이터 세그먼트로부터 요서 값들이 단조적으로 변화하는 특성을 이용하여 특성 벡터를 추출하고, 추출된 특성 벡터를 이용하여 공간 인덱스를 생성한다. 질의는 이 인덱스를 이용하여 (1) R-트리 여과, (2) 특성 여과, (3) 순서 여과, (4) 후처리의 네 단계로 처리된다. 다양한 실험을 통하여 제안된 기법의 효율성을 입증한다.

• 제어로봇시스템학회논문지
• /
• 제16권8호
• /
• pp.744-750
• /
• 2010

To segment instances of vehicle classes in a sparse stereo-matching data set, this paper presents an algorithm for clustering based on DP (Dynamic Programming). The algorithm is agglomerative: it begins with each element in the set as a separate cluster and merges them into successively larger clusters according to similarity of two clusters. Here, similarity is formulated as a cost function of DP. The proposed algorithm is proven to be effective by experiments performed on various images acquired by a moving vehicle.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 하계학술대회 논문집 G
• /
• pp.2243-2245
• /
• 1998

In this paper, we treat image segmentation as a graph partitioning problem. and use the normalized cut for segmenting the graph. The normalized cut criterion measures both the total dissimilarity between the different graphs as well as the total similarity within the groups. The minimization of this criterion can formulated as a generalized eigenvalues problem. We have applied this approach to segment static image. This criterion can be shown to be computed efficiently by a generalized eigenvalues problem

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2007년도 하계종합학술대회 논문집
• /
• pp.419-420
• /
• 2007

In this method, we extract peak periods using energy contents of each segment of music. This feature extraction method is equally applied on both the training and query music. Similarity matching algorithm is applied on the extracted feature values for identifying the query music from the database. The retrieval accuracy of 95% of our method is a pretty good result.

• 한국의류학회지
• /
• 제16권2호
• /
• pp.263-275
• /
• 1992

This study intends to provide strategic positioning of brand image analysed from the view point of perceptual dimensions of clothing consumers. Consumers are segmented on the basis of the attributes of brand image, and in each segment, perceptual map is composed according to multidimensional scaling. The results are as follows; 1. According to the Benefit Segmentation, it is statistically significant that the consumers are divided into 'product-factor oriented group 'and' image-factor oriented group'. 2. From the analysis of perceptual map upon the 'similarity of brand image,'image-factor oriented group 'perceives more differently than 'product-factor oriented group' 3. From the analysis of perceptual map with the evaluation of attributes of brand image, price, promotion and design are significant determinants in 'total consumer group'. In addition, store image is significant determinant in' image-factor oriented group' and quality is significant determinant in' product-factor oriented group'. According to the evaluation of consumers on 8 brands with determining attribute-vector, ranks of brands in each segment are similar in the vector of price and promotion but different in the vector of design between segment groups. 4. From the analysis of perceptual map upon the preference of brand image, the distribution of preference and position of ideal point are different between segment groups. 5. With evaluation of purchase habit, statistically significant differences are found between groups segmented in the degree of importance of attributes, purchasing motive, purchasing time, sources of information and expenses for clothes.

• 대한바이러스학회지
• /
• 제30권1호
• /
• pp.19-28
• /
• 2000

Two distinct hantaviruses have been isolated from Apodemus agrarius in 1976 and Rattus norvegicus in 1980 in Korea. Since our serosurveys conducted in 1994, a genetically distinct hantavirus from Apodemus peninsulae has been investigated. To isolate hantavirus from Apen insulae captured in Korea, the lung homogenate of seropositive Apeninsulae inoculated Vero E6 cells. Viral antigen was detected in a progressively higher percentage of cells with subsequent passage after 80 days postinoculation. The new isolate from seropositive Apodemus peninsulae was designated Suchong virus after Suchong valley located in northeastern region of South Korea. Comparing with hantaan virus 76-118 strain, Suchong virus-1, 2, 3 and 4 showed the similarity of $71.0{\sim}91.8%$ at nucleotide and $90.9{\sim}94.8%$ at amino acid sequences in 231 nucleotides region of M segment, and the similarity of $75.1{\sim}81.0%$ at nucleotide and $97.5{\sim}100%$ at amino acid sequences in 237 nucleotides of S segment.

• 정보처리학회논문지D
• /
• 제10D권2호
• /
• pp.283-292
• /
• 2003

연속된 일차원 실수로 이루어진 시계열 데이터는 데이터 마이닝이나 데이터 웨어하우징과 같은 다양한 데이터베이스 응용 분야에서 연구되어져 왔다. 그러나 최근의 복잡한 비즈니스 환경에서, 다차원 데이터 시퀀스(multidimensional data sequence : MDS)는 일차원 시계열 데이터와 더불어 그 중요성이 더해가고 있다. 다차원 데이터 시퀀스의 예로써, 비디오 스트림은 색상과 질감 등의 속성들로 이루어진 다차원 공간상에서 MDS로 나타낼 수 있다. 본 논문에서는 패턴 유사성 검색에서 사용되는 효과적인 유사성 척도를 제시한다. 하나의 MDS는 여러 개의 세그먼트(segment)로 나누어지며, 각 세그먼트는 다양한 의미적인 특징들로 표현된다. 유사성 척도는 이러한 세그먼트에 대해서 정의되는데 이 척도를 사용하여 어떤 주어진 질의 시퀀스에 대하여 무관한 세그먼트들은 검색 대상에서 일차적으로 제외된다. 데이터 시퀀스와 질의 시퀀스 모두 세그먼트 단위로 분할되며, 질의 처리는 전체 시퀀스의 모든 데이터를 검색하지 않고 데이터 세그먼트와 질의 세그먼트의 특징을 비교하는 것을 기초로 하여 수행된다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.877-881
• /
• 2005

In this paper, a novel method for analyzing a textile fabric structure is proposed to segment each yarn of the textile fabric from voxel data made out of its X-ray computed tomography (CT) images. In order to segment the each yarn, directions of fibers, of which yarn consists, are firstly estimated by correlating the voxel with a fiber model. Second, each fiber is reconstructed by clustering the voxel of the fiber using the estimated fiber direction as a similarity. Then, each yarn is reconstructed by clustering the reconstructed fibers using a distance which is newly defined as a dissimilarity. Consequently, each yarn of the textile fabric is segmented from the voxel data. The effectiveness of the proposed method is confirmed by experimentally applying the method to voxel data of a sample plain woven fabric, which is made of polyester two folded yarn. The each two folded yarn is correctly segmented by the proposed method.

• 한국기계가공학회지
• /
• 제13권1호
• /
• pp.30-38
• /
• 2014

The purpose of this study is to build and manage environment models with line segments from sonar range data on obstacles in unknown and varied environments. The proposed method therefore employs a two-stage data-transform process in order to extract environmental line segments from range data on obstacles. In the first stage, the occupancy grid extracted from the range data is accumulated to form a two-dimensional local histogram grid. In the second stage, a line histogram extracted from a local histogram grid is based on a Hough transform, and matching serves as a means of comparing each of the segments on a global line segments map against the line segments to detect the degree of similarity in the overlap, orientation, and arrangement. Each of these tests is formulated by comparing one of the parameters in the segment representation. After the tests, new line segments can be found at maximum-density cells in the line histogram, and they are composed onto the global line segment map. The proposed technique is demonstrated in experiments in an indoor environment.

• 대한바이러스학회지
• /
• 제29권1호
• /
• pp.23-31
• /
• 1999

Hantavirus is a genus of the Bunyaviridae family causing two serious diseases, hemorrhagic fever with renal syndrome (HFRS) and hantavirus pulmonary syndrome (HPS). Puumala virus is a member of hantavirus originally found in Europe, and its natural reservoir is Clethrionomys glareolus. It is also associated with the human disease nephropathia epidemica, a milder form of HFRS. To identify the hantaviruses in bats, bats were collected from Jeong-Sun, Won-Joo, Chung-Ju and Hwa-Cheon area in Korea, and nested RT-PCR was performed with serotype specific primer from M segment. Interestingly, Puumala virus was detected in bats (Rhinolophus ferrum-equinum) only from Won-Joo. The 327 bp nested RT-PCR product, was sequenced. The sequence database search indicates that the sequence is homologous to the published sequence of Puumala viruses. The sequence similarities were ranged from 71% to 97%. The highest sequence similarity was 97% with Puumala virus Vranicam strain, and the lowest was 71% with Puumala virus K27 isolate. Puumala virus Vranicam strain was isolated from a bank vole (Clethrionomys glareolus) in Bosnia-Hercegovina. Puumala virus K27 was isolated from human in Russia. This analysis confirms that bats (Rhinolophus ferrum-equinum) in Korea are natural reservoir of Puumala virus.