• The KIPS Transactions:PartD
• /
• v.8D no.6
• /
• pp.813-822
• /
• 2001

In this paper we proposed four data migration methods based on time segmented storage structure including past segment, current segment, and future segment. The migration methods proposed in this paper are the Time Granularity migration method, the LST-GET (Least valid Start Time-Greatest valid End Time) migration method, the AST-AET (Average valid Start Time-Average valid End Time) migration method, and the Min-Overlap migration method. In the each data migration method we define the dividing criterion among segments and entity versions to store on each segment. We measured the response time of queries for the proposed migration methods. When there are no LLTs (Long Lived Tuples), the average response time of AST-AET migration method and LST-GET migration method are smaller than that of Time Granularity migration method. In case of existing LLT, the performance of the LST-GET migration method decreased. The AST-AET migration method resulted in better performance for queries than the Time Granularity migration method and the LST-GET migration method. The Min-Overlap migration method resulted in the almost equal performance for queries compared with the AST-AET migration method, in case of storage utilization more efficient than the AST-AET.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06c
• /
• pp.183-185
• /
• 2012

행위 DB로부터 행위패턴 분석 및 마이닝을 위해서는 정교한 행위패턴 모델링 기술이 수반되어야 한다. 기존의 그래프기반 행위 패턴 모델링 방법은 하루 행위 시퀀스들의 동일한 행위 시퀀스 세그먼트를 찾아 하나의 행위 시퀀스로 결합시켜 행위 그래프를 생성하였다. 이 방법은 서로 다른 시간에 발생한 행위 시퀀스 세그먼트들이 하나의 행위 시퀀스로 결합되는 문제가 발생한다. 본 논문에서는 하루의 행위 시퀀스를 시간 세그먼트 단위로 분할하고, 각 시간 세그먼트별로 행위 그래프를 생성하여 정교한 행위 그래프 모델을 수립하는 방법을 제안한다. 그래프 마이닝 기법들을 활용한 실험을 통하여 제안하는 행위패턴 모델링 기법이 기존의 행위 그래프 모델 기법보다 더 유용함을 보인다.

• Journal of Korea Multimedia Society
• /
• v.5 no.2
• /
• pp.158-166
• /
• 2002

In this paper, we proposed a contour shape description method which use the CFR(contour fluctuation ratio) feature. The CFR is the ratio of the line length to the curve length of a contour segment. The line length means the distance of two end points on a contour segment, and the curve length means the sum of distance of all adjacent two points on a contour segment. We should acquire rotation and scale invariant contour segments because each CFR is computed from contour segments. By using the interleaved contour segment of which length is proportion to the entire contour length and which is generated from all the points on contour, we could acquire rotation and scale invariant contour segments. The CFR can describes the local or global feature of contour shape according to the unit length of contour segment. Therefore we describe the shape of objects with the feature vector which represents the distribution of CFRs, and calculate the similarity by comparing the feature vector of corresponding unit length segments. We implemented the proposed method and experimented with rotated and scaled 165 fish images of fifteen types. The experimental result shows that the proposed method is not only invariant to rotation and scale but also superior to NCCH and TRP method in the clustering power.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.12
• /
• pp.963-967
• /
• 2009

The Log-structured File System (LFS) collects all modified data into a memory buffer and writes them sequentially to a segment on disk. Therefore, it has the potential to utilize the maximum bandwidth of storage devices where sequential writes are much faster than random writes. However, as disk space is finite, LFS has to conduct cleaning to produce free segments. This cleaning operation is the main reason LFS performance deteriorates when file system utilization is high. To overcome painful cleaning and reduced performance of LFS, we propose the segment space recycling (SSR) scheme that directly writes modified data to invalid areas of the segments and describe the classification method of data and segment to consider locality of reference for optimizing SSR scheme. We implement U-LFS, which employs our segment space recycling scheme in LFS, and experimental results show that SSR scheme increases performance of WOLF by up to 1.9 times in HDD and 1.6 times in SSD when file system utilization is high.

• Phonetics and Speech Sciences
• /
• v.11 no.3
• /
• pp.39-47
• /
• 2019

We propose a method to improve the performance of spontaneous speech recognizers by extending their phone set using speech data. In the proposed method, we first extract variable-length phoneme-level segments from broadcast speech signals, and convert them to fixed-length latent vectors using an long short-term memory (LSTM) classifier. We then cluster acoustically similar latent vectors and build a new phone set by choosing the number of clusters with the lowest Davies-Bouldin index. We also update the lexicon of the speech recognizer by choosing the pronunciation sequence of each word with the highest conditional probability. In order to analyze the acoustic characteristics of the new phone set, we visualize its spectral patterns and segment duration. Through speech recognition experiments using a larger training data set than our own previous work, we confirm that the new phone set yields better performance than the conventional phoneme-based and grapheme-based units in both spontaneous speech recognition and read speech recognition.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.415-416
• /
• 2010

이 논문에서 우리는 텍스쳐 정보가 적은 영역이 존재하는 입력 영상들에 대해서도 안정적인 복원을 도출하는 새로운 다시 점 스테레오 방법을 제시한다. 제안된 방법에서는 입력 영상들을 인접한 픽셀간의 색 유사성을 이용하여 세그먼테이션한 후, 세그먼트 단위로 다시점 스테레오를 수행한다. 특히 그 과정에서 한 영상 내의 이웃한 세그먼트들의 깊이 값 유사성, 그리고 서로 다른 시점에서 상응하는 세그먼트 간의 깊이 값 일관성을 가정하여, 텍스쳐 정보가 적은 영역에 대해서도 안정적으로 3D 점들을 생성해준다. 생성된 3D 점들은 그래프 컷 기반의 복원 알고리즘을 통해 일관된 3D 표면으로 복원 되었고, 복원 결과는 제안된 방법이 실제로 기존의 다른 다시점 스테레오에 비해 보다 안정적으로 깊이 정보를 추출할 수 있음을 보여준다. 결과적으로 제안된 방법은 보다 일반적이고 실생활에 가까운 입력 영상들에 대해서도 3D 복원을 수행할 수 있는 방향을 제시한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.265-268
• /
• 2022

Video-based Point Cloud Compression(V-PCC) 부호화기의 세그먼트 정제(Refining segmentation) 과정은 3D 세그먼트를 2D 패치 데이터로 효율적으로 변환하기 위한 V-PCC 부호화기의 핵심 파트이지만, 많은 연산량을 필요로 하는 모듈이다. 때문에 이미 TMC2 에 Fast Grid-based refine segmentation 과정이 구현되어 있으나, 아직도 세그먼트 정제 기술의 연산량은 매우 높은 편이다. 본 논문에서는 현재 TMC2 에 구현되어 있는 Fast Gridbased Refine Segmentation 을 살펴보고, 복셀(Voxel) 타입에 따른 특성에 맞춰 두 가지 조건을 추가하는 고속화 알고리즘을 제안한다. 실험 결과 압축성능(BD-BR)은 TMC2 와 거의 차이를 보이지 않았지만, 모듈 단위 평균 10% 연산량이 절감되는 것을 확인하였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.11C
• /
• pp.1046-1054
• /
• 2002

In this paper, we propose a subnetwork-based segment restoration scheme to reduce the restoration time and restoration resources. And we compare and analyze the restoration performance according to the size of divided subnetworks. Segment restoration is based on network partitioning where a large network is divided into several small subnetworks and the end-to-end data path is divided into multiple segments according to the subnetworks. In segment restoration, the link/node failure is restored by segment instead of end-to-end path. Because most faults are restored within the subnetwork, the restoration performance can be improved. From the simulation analysis, we verified that the proposed segment restoration has advantage of restoration time and backup resource utilization.

• The KIPS Transactions:PartD
• /
• v.10D no.2
• /
• pp.283-292
• /
• 2003

One-dimensional time-series data have been studied in various database applications such as data mining and data warehousing. However, in the current complex business environment, multidimensional data sequences (MDS') become increasingly important in addition to one-dimensional time-series data. For example, a video stream can be modeled as an MDS in the multidimensional space with respect to color and texture attributes. In this paper, we propose the effective similarity measures on which the similar pattern retrieval is based. An MDS is partitioned into segments, each of which is represented by various geometric and semantic features. The similarity measures are defined on the basis of these segments. Using the measures, irrelevant segments are pruned from a database with respect to a given query. Both data sequences and query sequences are partitioned into segments, and the query processing is based upon the comparison of the features between data and query segments, instead of scanning all data elements of entire sequences.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.298-300
• /
• 2004

바이오 인포메틱스에서의 데이터 검색은 DNA와 단백질 시퀀스에 대해서 주로 이루어지며, 지금까지의 연구는 주로 DNA와 단백질 1차 구조의 검색에 대해 이루어졌다. 단백질 2차구조는 1차구조 내 인접한 아미노산들의 공간적인 배열을 나타내며. 단백질의 기능을 예측하는데 중요한 3차구조의 지역적 아미노산의 특성을 나타낸다. 따라서 2차구조에 대한 검색은 단백질의 기능을 이해하는데 매우 중요한 역할을 한다［1］. 이 논문에서는 단백질 2차구조 및 질의 문자열을 세그먼트 단위로 나누고 검색하는 r41의 방법을 개선하여 세그먼트를 조합한 클러스터 구조 및 Look Ahead를 사용해 Exact Matching 및 Wildcard Matching 질의를 효율적으로 처리할 수 있는 기법을 제시한다.