• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.05c
• /
• pp.272-276
• /
• 2002

본 논문은 영역을 병합할 때 두 영역의 색상 차를 영역 병합의 제한 조건으로 사용하는 칼라 영상 분할 기법을 제안하였다. 이는 먼저 영역의 경계선 정보를 잘 보존하기 위해서 RGB 공간상에서 수리형태학 필터와 변형된 워터쉐드 알고리즘을 이용하여 칼라 영상을 과분할 한다. 그리고 영역 간의 색상 차를 제한 조건으로 사용하는 영역 병합 과정을 반복 수행하여 칼라 영상의 분할 결과를 얻는다. 이는 인간 시각 시스템이 색상, 채도, 명도의 형태로 색을 구분하는 것을 기반으로 한다. 명도가 낮지 않는 경우에 색차 보다 색상 차가 중요한 요소로 작용하기 때문에 이를 영역 병합의 제한 조건으로 사용한다. 실험결과에서 제안된 칼라 영상 분할 기법은 다양한 칼라 영상에 대하여 적은 개수의 영역으로 동일한 색상을 가지는 영역의 경계선을 유지하는 효율적인 분할을 보임을 확인하였다.

• Annual Conference of KIPS
• /
• 2003.11c
• /
• pp.1395-1398
• /
• 2003

최근 위치기반 기술의 급속한 발전으로 인하여 이동 객체를 효율적으로 관리하기 위한 색인 구조의 필요성이 증가하고 있다. 본 논문에서는 KDB-트리를 기반으로 하는 새로운 형태의 시공간 색인 구조인 TPKDB-트리 (Time Parameterized KDB-Tree)를 제안한다. 제안하는 색인 구조는 갱신 비용을 최소화 하여 이동 객체 검색의 효율성을 증가시키고 노드 내에 포함되어 있는 이동 객체의 변화를 시간에 대한 파라미터로 유지함으로서 효율적으로 이동 객체의 미래 위치 검색을 지원한다. 또한, 공간활용도를 최대화하기 위해 EFP 분할 (Enhanced First Division Splitting) 기법을 제안한다. 제안하는 색인 구조의 우수성을 입증하기 위해 실험을 통해 다른 색인 구조차의 성능 비교를 수행한다.

• The Journal of Korean Association of Computer Education
• /
• v.6 no.4
• /
• pp.49-59
• /
• 2003

In order to index exactly moving objects(vehicle, mobile phone, PDA, etc.) in a mobile database, continuous updates of their locations arc inevitable as well as time-consuming. In this paper, we propose a spatial indexing method of moving objects for the efficient mobile map services. In detail, we analyze the characteristics of both the method to re-index all the objects after each time period and the method to update immediately the locations on reporting their locations. We also newly propose a bucket split method using the properties of moving objects in order to minimize the number of database updates. The experiments conducted on the environments of moving object show that the proposed indexing method is appropriate to map services for mobile devices.

• Journal of Korea Spatial Information System Society
• /
• v.6 no.1 s.11
• /
• pp.19-29
• /
• 2004

Recently, studies have been performed to improve the cache performance of the R-Tree in main memory. A general mothed to improve the cache performance of the R-Tree is to reduce size of an entry so that a node can store more entries and fanout of it can increase. However, this method generally requites additional process to reduce information of entries and do not support incremental updates. In addition, the cache miss always occurs on moving between a parent node and a child node. To solve these problems efficiently, this paper proposes and evaluates the PR-Tree that is an extended R-Tree indexing method using prefetching in main memory. The PR-Tree can produce a wider node to optimize prefetching without additional modifications on the R-Tree. Moreover, the PR-Tree reduces cache miss rates that occur in moving between a parent node and a child node. In our simulation, the search performance, the update performance, and the node split performance of the PR-Tree improve up to 38%. 30%, and 67% respectively, compared with the original R-Tree.

• Annual Conference of KIPS
• /
• 2009.04a
• /
• pp.348-351
• /
• 2009

유비쿼터스 사회를 실현하는 핵심기술인 u-GIS 공간정보 기술은 데이터 스트림 처리 시스템(Data Stream Management System)과 지리정보 시스템(Geography Information System)이 결합된 플랫폼인 u-GIS DSMS를 요구한다. u-GIS DSMS는 GeoSeonsor에서 수집되는 센서 테이터와 GIS의 공간정보 데이터를 결합하여 처리하는 공간영역질의가 다수 요구된다. 이런 공간영역질의들은 특정 지역에 밀집하게 등록되는 경향이 있으며, 유사한 프리디킷을 가질 가능성이 높다. 이러한 특징은 공간영역질의가 특정 지역에 밀집되면 다수의 비슷한 연산들이 반복적으로 처리하기 때문에 시스템 성능이 저하 될 것이다. 이를 해결하기 위해 영역질의 색인기법 연구가 활발히 진행되고 있다. 그러나 기존의 VCR-Index와 CQI-Index 기법은 질의영역을 셀 구조나 가상구조로 분할하여 처리하기 때문에 자원 및 연산을 공유 할 수 없어 질의 처리 속도가 현저히 저하되기 때문에 대량의 공간영역질의 처리에는 부적합하다. 그래서 본 논문에서는 공간영역질의의 효율적인 연산 공유를 위한 질의영역 밀집도 기반의 그룹화 기법을 제안한다. 이 기법은 질의영역의 밀집도를 이용하여 공간영역질의들을 그룹화 후 색인을 구성한다. 색인된 영역들의 데이터는 단일 큐로 구성 후 질의들의 프리디킷을 분석하여 자원 및 연산 공유기법을 통해 기존의 기법보다 처리 속도 향상 및 메모리 사용을 감소시켰다.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.9
• /
• pp.335-344
• /
• 2018

The power model is needed to handle the power consumption of mobile AMOLED display at the software level. However, the existing studies to generate the power model have required the experimental environment and equipment for the power measurement activity. In addition, the combination of RGB values used for modeling was imprudent and small, so it was difficult to reflect the mutual influence between the RGB values into the model. To solve these problems, we propose an RGB color space partitioning method, which is used to prudently sample the combinations of the RGB values based on the color or the power. We also propose a process for generating a mapping table composed of . We analyzed the characteristics of the samples generated according to the proposed partitioning methods, taking into account the color and the power, and generated the mapping table for the AMOLED display. Furthermore, we confirmed the reusability of the mapping table by utilizing one mapping table multiple times in evaluating different power models. These mapping tables are provided to researchers and can be used to generate and evaluate power models without power measurement activities.

• The KIPS Transactions:PartD
• /
• v.8D no.4
• /
• pp.305-312
• /
• 2001

Recently, high-dimensional index structures have been required for similarity search in such database applications s multimedia database and data warehousing. In this paper, we propose a new cell-based signature tree, called CS-tree, which supports efficient storage and retrieval on high-dimensional feature vectors. The proposed CS-tree partitions a high-dimensional feature space into a group of cells and represents a feature vector as its corresponding cell signature. By using cell signatures rather than real feature vectors, it is possible to reduce the height of our CS-tree, leading to efficient retrieval performance. In addition, we present a similarity search algorithm for efficiently pruning the search space based on cells. Finally, we compare the performance of our CS-tree with that of the X-tree being considered as an efficient high-dimensional index structure, in terms of insertion time, retrieval time for a k-nearest neighbor query, and storage overhead. It is shown from experimental results that our CS-tree is better on retrieval performance than the X-tree.

• Journal of KIISE:Databases
• /
• v.34 no.1
• /
• pp.59-69
• /
• 2007

The trajectory of a tag in RFID system is represented as a interval that connects two spatiotemporal locations captured when the tag enters and leaves the vicinity of a reader. Whole trajectories of a tag are represented as a set of unconnected interval because the location of the tag which left the vicinity of a reader is unknown until it enters the vicinity of another reader. The problems are that trajectories of a tag are not connected. It takes a long time to find trajectories of a tag because it leads to searching the whole index. To solve this problem, we propose a technique that links two intervals of the tag and an index scheme called SLR-tree. We also propose a sharing technique of link information between two intervals which enhances space utilization of nodes, and propose a split policy that preserves shared-link information. And finally, we evaluate the performance of the proposed index and prove that the index processes history queries efficiently.

• The KIPS Transactions:PartD
• /
• v.10D no.4
• /
• pp.591-602
• /
• 2003

Delay and discontinuance phenomenon of service are cause by sudden increase of the network communication amount and the quantity consumed of resources when Internet users are driven excessively to a conventional single large database sewer. To solve these problems, spatial database cluster consisted of several single nodes on high-speed network to offer high-performance is risen. But, research about spatial join operation that can reduce the performance of whole system in case process at single node is not achieved. So, in this paper, we propose efficient parallel spatial join processing method in a spatial database cluster system that uses data partitions and replications method that considers the characteristics of space data. Since proposed method does not need the creation step and the assignment step of tasks, and does not occur additional message transmission between cluster nodes that appear in existent parallel spatial join method, it shows performance improvement of 23% than the conventional parallel R-tree spatial join for a shared-nothing architecture about expensive spatial join queries. Also, It can minimize the response time to user because it removes redundant refinement operation at each cluster node.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.2
• /
• pp.69-78
• /
• 2002

This paper addresses a method to automatically detect out a person's face from a given image that consists of a hair and face view of the person and a complex background scene. Out method involves an effective detection algorithm that exploits the spatial distribution characteristics of human skin color via an adaptive fuzzy color classifier (AFCC), The universal skin-color map is derived on the chrominance component of human skin color in Cb, Cr and their corresponding luminance. The desired fuzzy system is applied to decide the skin color regions and those that are not. We use RGB model for extracting the hair color regions because the hair regions often show low brightness and chromaticity estimation of low brightness color is not stable. After some preprocessing, we apply convex-hull to each region. Consequent face detection is made from the relationship between a face's convex-hull and a head's convex-hull. The algorithm using the convex-hull shows better performance than the algorithm using pattern method. The performance of the proposed algorithm is shown by experiment. Experimental results show that the proposed algorithm successfully and efficiently detects the faces without constrained input conditions in color images.