• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 1999.03a
• /
• pp.347-350
• /
• 1999

자율무인잠수정은 자율운항을 위해서 자동화된 제어시스템이 필요하다. 제어시스템은 기능적 측면에서 임무계획단계(mission planning level), 임무제어단계(mission control level), 선체제어단계(vehicle control level)로 구분한다. 자율무인잠수정의 효과적인 임무 수행을 위해서는 임무제어단계의 운행 경로 설정과 제어가 중요하다. 자율무인잠수정은 잠수정의 주변환경을 추상화한 후 탐색기법을 이용하여 경로를 설정한다. 이때 검색기법의 효율적 적용을 위해서는 효과적으로 추상화된 탐색모형이 필요하다. 대표적인 탐색모형으로는 3차원 격자절대좌표 모형(3-dimensional grid unit coordinate model)[1]을 들 수 있다. 그러나 이 모형은 불필요한 동작의 반복, 이동 격자에 따른 비일관성과 같은 취약점이 존재한다. 본 연구에서는 이 모형의 취약점을 개선하기 위해서 자유무인잠수정의 위치 기반 상대적 격자좌표 모형(relative grid unit coordinate model based on AUV state)을 제안한다.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.7
• /
• pp.1724-1737
• /
• 1999

Multimedia data service applications have to store and manipulate LOBs(unstructured large objects) composing multimedia data. As such, the tertiary storage devices such as an optical disk jukebox and a tape library that consist of a number of platters (the disks in case of an optical disk jukebox and the cartridge tapes in case of a tape library) have been considered essential for the storage system of a DBMS in order to efficiently support storage and management of vary large volume of data. Since the latency with tertiary storage is too long, the schemes for efficient retrieval of LOBs out of tertiary storage need to be investigated. In this paper, we investigated the tertiary I/O Considering the performance characteristics of the LOBs, we proposed various I/O scheduling heuristic algorithms that reduce latency in query processing with LOB retrieval from tertiary storage, and evaluated their performance through a detailed simulation.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10a
• /
• pp.293-296
• /
• 2000

본 논문에서는 주제어 인식에 기반한 문서영상의 검색 및 색인 시스템에 적용하기 위한 단어 단위 분한 시스템을 제안한다. 제안 시스템은 영상 전처리, 문서 구조 분석을 통해 추출된 텍스트 영역을 입력으로 단어 단위 분할을 수행하는데, 텍스트 영역에 대해 텍스트 라인을 분할하고 분할된 텍스트 라인을 단어 단위로 분할하는 계층적 접근 방법을 사용한다. 텍스트라인 분할은 수평 방향 투영 프로파일을 적용하여 분할 지점을 구한다. 그리고 단어 분할은 연결요소들을 추출한 후 연결요소간의 gap 정보를 구하고, gap 군집화 기법을 사용하여 단어 단위 분한 지점을 구한다. 이때 단어 단위 분할의 성능을 저하시키는 특수기호에 대해서는 휴리스틱 정보를 이용하여 검출한다. 제안 시스템의 성능 평가는 50개의 텍스트 영역에 적용하여 99.83%의 정확도를 얻을 수 있었다.

• The KIPS Transactions:PartB
• /
• v.8B no.6
• /
• pp.675-684
• /
• 2001

This paper describes a method for stemming of Korean language by using unsupervised learning from raw corpus. This technique does not require a lexicon or any language-specific knowledge. Since we use unsupervised learning, the time and effort required for learning is negligible. Unlike heuristic approaches that are theoretically ungrounded, this method is based on widely accepted statistical methods, and therefore can be easily extended. The method is currently applied only to Korean language, but it can easily be adapted to other agglutinative languages, since it is not language-dependent.

• Journal of KIISE:Software and Applications
• /
• v.30 no.7_8
• /
• pp.610-617
• /
• 2003

Similarity matching in video database is of growing importance in many new applications such as video clustering and digital video libraries. In order to provide efficient access to relevant data in large databases, there have been many research efforts in video indexing with diverse spatial and temporal features. however, most of the previous works relied on sequential matching methods or memory-based inverted file techniques, thus making them unsuitable for a large volume of video databases. In order to resolve this problem, this paper proposes an effective and scalable indexing technique using a trie, originally proposed for string matching, as an index structure. For building an index, we convert each frame into a symbol sequence using a window order heuristic and build a disk-resident trie from a set of symbol sequences. For query processing, we perform a depth-first search on the trie and execute a temporal segmentation. To verify the superiority of our approach, we perform several experiments with real and synthetic data sets. The results reveal that our approach consistently outperforms the sequential scan method, and the performance gain is maintained even with a large volume of video databases.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.5
• /
• pp.175-182
• /
• 2010

The paper presents a novel algorithm for face and iris detection with the application for driver iris monitoring. The proposed algorithm consists of the following major steps: Skin-color segmentation, facial features segmentation, and iris positioning. For the skin-segmentation we applied a multi-layer perceptron to approximate the statistical probability of certain skin-colors, and filter out those with low probabilities. The next step segments the face region into the following categories: eye, mouth, eye brow, and remaining facial regions. For this purpose we propose a novel segmentation technique based on estimation of facial class probability density functions (PDF). Each facial class PDF is estimated on the basis of salient features extracted from a corresponding facial image region. Then pixels are classified according to the highest probability selected from four estimated PDFs. The final step applies the circular Hough transform to the detected eye regions to extract the position and radius of the iris. We tested our system on two data sets. The first one is obtained from the Web and contains faces under different illuminations. The second dataset was collected by us. It contains images obtained from video sequences recorded by a CCD camera while a driver was driving a car. The experimental results are presented, showing high detection rates.

• The KIPS Transactions:PartD
• /
• v.11D no.5
• /
• pp.1021-1030
• /
• 2004

When any data-centric XML documents are stored and managed in RDBMS, schema extracting from XML Schema is an imminent problem. Furthermore, when they are stored in partitioned way on created table, lots of null values will be produced and/or be increased cost for join, so we need a solution to solve these problems. This paper proposes a Schema Hybrid Inlining technique to generate relational schema of XML documents. The suggested technique creates a relational schema based on the XML Schema graph. Also, the technique expands the legacy Inlining method by manual and heuristic processing table generation method of cardinality and in-degree. Then, an ancestor node, terminal node, and creating a new table on the derived relation in this technique. DTD-dependent storage method uses partitioned storing and relation path element information reformed structured-searching method to reduce joining operation cost that is a weak point of it.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.4B
• /
• pp.236-244
• /
• 2012

The minimum energy broadcast problem is for all deployed nodes to minimize a total transmission energy for performing a broadcast operation in wireless networks. In this paper, we propose a Tabu search algorithm to solve efficiently the minimum energy broadcast problem on the basis of meta-heuristic approach in wireless sensor networks. In order to make a search more efficient, we propose a novel neighborhood generating method and a repair function of the proposed algorithm. We compare the performance of the proposed algorithm with other existing algorithms through some experiments in terms of the total transmission energy of nodes and algorithm computation time. Experimental results show that the proposed algorithm is efficient for the minimum energy broadcast problem in wireless sensor networks.

• Journal of the Korea Convergence Society
• /
• v.9 no.7
• /
• pp.1-7
• /
• 2018

Feature selection, one of data preprocessing techniques, is one of major research areas in many applications dealing with large dataset. It has been used in pattern recognition, machine learning and data mining, and is now widely applied in a variety of fields such as text classification, image retrieval, intrusion detection and genome analysis. The proposed method is based on a genetic algorithm which is one of meta-heuristic algorithms. There are two methods of finding feature subsets: a filter method and a wrapper method. In this study, we use a wrapper method, which evaluates feature subsets using a real classifier, to find an optimal feature subset. The training dataset used in the experiment has a severe class imbalance and it is difficult to improve classification performance for rare classes. After preprocessing the training dataset with SMOTE, we select features and evaluate them with various machine learning algorithms.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.12
• /
• pp.2221-2229
• /
• 2006

Optimal path search algorithm which is a killer application of mobile device to utilize location information should consider traffic flows of the roads as well as the distance between a departure and destination. The existing path search algorithms, however, are net able to cope efficiently with the change of the traffic flows. In this paper, we propose a new optimal path search algorithm. The algorithm takes the current flows into consideration in order to reduce the cost to get destination. It decomposes the road network into Fixed Grid to get variable heuristics. We also carry out the experiments with Dijkstra and Ar algorithm in terms of the execution time, the number of node accesses and the accuracy of path. The results obtained from the experimental tests show the proposed algorithm outperforms the others. The algorithm is highly expected to be useful in a advanced telematics systems.