• The Transactions of the Korea Information Processing Society
• /
• v.6 no.2
• /
• pp.543-551
• /
• 1999

The primary purposes in this paper are to analyze the class separation of features in handwritten numeral recognition and to make use of the results in feature selection. Using the Parzen window technique, we compute the class distributions and define the class separation to be the overlapping distance of two class distributions. The dimension of a feature vector is reduced by removing the void or redundant feature cells based on the class separation information. The experiments have been performed on the CENPARMI handwritten numeral database, and partial classification and full classification have been tested. The results show that the class separation is very effective for the feature selection in the 10-class handwritten numeral recognition problem since we could reduce the dimension of the original 256-dimensional feature vector by 22%.

• Journal of Korea Multimedia Society
• /
• v.10 no.5
• /
• pp.583-593
• /
• 2007

We study the page replacement algorithms for demand paging, called CFLRU/C, CFLRU/E and DL-CFLRU/E, that reduce the number of erase operations and improve the wear-leveling degree of flash memory. Under the CFLRU/C and CFLRU/E algorithms, the victim page is the least recently used dean page within the pre-specified window. However, when there is not any dean page within the window, the CFLRU/C evicts the dirty page with the lowest frequency while the CFLRU/E evicts the dirty page with the highest number of erase operations. The DL-CFLRU/E algorithm maintains two page lists called the dean page list and the dirty page list, and first finds the page within the dean page list when it selects a victim. However, when it can not find any dean page within the dean page list, it evicts the dirty page with the highest number of erase operations within the window of the dirty page list. In this thesis, we show through simulation that the proposed schemes reduce the number of erase operations and improve the wear-leveling than the existing schemes like LRU.

• The KIPS Transactions:PartD
• /
• v.12D no.3 s.99
• /
• pp.345-354
• /
• 2005

Subsequence matching is one of the most important operations in the field of data mining. The existing subsequence matching algorithms use only one index, and their performance gets worse as the difference between the length of a query sequence and the site of windows, which are subsequences of a same length extracted from data sequences to construct the index, increases. In this paper, we propose a new subsequence matching method based on index interpolation to overcome such a problem. An index interpolation method constructs two or more indexes, and performs search ing by selecting the most appropriate index among them according to the given query sequence length. In this paper, we first examine the performance trend with the difference between the query sequence length and the window size through preliminary experiments, and formulate a search cost model that reflects the distribution of query sequence lengths in the view point of the physical database design. Next, we propose a new subsequence matching method based on the index interpolation to improve search performance. We also present an algorithm based on the search cost formula mentioned above to construct optimal indexes to get better search performance. Finally, we verify the superiority of the proposed method through a series of experiments using real and synthesized data sets.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.45-55
• /
• 2021

This paper has proposed an algorithm that detecting for dense small vehicle in large image efficiently. It is consisted of two Ensemble Deep-Learning Network algorithms based on Coarse to Fine method. The system can detect vehicle exactly on selected sub image. In the Coarse step, it can make Voting Space using the result of various Deep-Learning Network individually. To select sub-region, it makes Voting Map by to combine each Voting Space. In the Fine step, the sub-region selected in the Coarse step is transferred to final Deep-Learning Network. The sub-region can be defined by using dynamic windows. In this paper, pre-defined mapping table has used to define dynamic windows for perspective road image. Identity judgment of vehicle moving on each sub-region is determined by closest center point of bottom of the detected vehicle's box information. And it is tracked by vehicle's box information on the continuous images. The proposed algorithm has evaluated for performance of detection and cost in real time using day and night images captured by CCTV on the road.

• Korean Journal of Materials Research
• /
• v.9 no.3
• /
• pp.257-262
• /
• 1999

The selective area growth of GaN by metal organic chemical vapor deposition has been carried out on GaN/ sapphire substrate using $SiO_2$ mask. We investgated the effect of growth parameters such as flow rate of $NH_3$(500­~1300sccm) and the growth temperature(TEX>$950~1060^{\circ}C$) on the growth selectivity and characteristics of GaN using the Scanning Electron Microscopy(SEM). The selectivity of GaN improved as flow rate of NH, and growth temperature in­creased. But the grown GaN shapes on the substrate windows was independent of the flow rate of $NH_3$. On the pattern shapes such as circle, stripe, and radiational pattern(rotate the stripe pattern by $30^{\circ}, 45^{\circ}$), we observed the hexagonal pyramid, the lateral over growth on the mask layer, and the difference of the lateral growth rate depending on growth condition.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.1
• /
• pp.165-171
• /
• 2005

In this paper, the centralized dynamic load-balancing was processed effectively by using the network based parallel genetic algorithm. Unlike the existing method using genetic algorithm, the performance of central scheduler was improved by distributing the process for the searching of the optimal task assignment to clients. A roulette wheel selection and an elite preservation strategy were used as selection operation to improve the convergence speed of optimal solution. A chromosome was encoded by using sliding window method. And a cyclic crossover was used as crossover operation. By the result of simulation for the performance estimation of central scheduler according to the change of flexibility of load-balancing method, it was verified that the performance is improved in the proposed method.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2002.11a
• /
• pp.512-519
• /
• 2002

In this paper, we implemented a loom component to be placed in a window and a looms management system which is able to connect database and to process various queries. The implemented system has aggregate query processing functions for the loom components existing in the selected area by the mouse and it also has high level query processing functions represented with chart and pivot table; it can be used as a decision support system. The proposed system can detect temporal or persistent problems of the Inn. Therefore, it can be used to raise the productivity and to reduce the cost in textile companies by coping with the situation properly.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06e
• /
• pp.335-338
• /
• 1998

본 논문에서는 음성인식 알고리즘인 HMM을 사용하여 Windows 환경에서 온라인으로 사용할 수 있는 음성인식 게임“Voice Illust Magic”개발에 관하여 소개한다. 사용자와 컴퓨터가 상호작용(Interaction)할 수 있는 매체를 마우스와 키보드뿐만 아니라 게임에 필요한 명령어를 음성인식으로 실행함으로써 정보전달이 매우 효과적으로 이루어져 사용자가 접근하기 쉽고 편리하게 되었으며 의사전달 효율을 높일 수 있었다. 음성인식 과정을 온라인으로 마이크를 통해 들어온 음성을 자동으로 끝점을 검출한 후, Mel-Cepstrum을 추출하여 Word 단위의 reference HMM과 비교하여 최적의 model이 선택되면, 윈도우즈에게 메시지를 보내어 마우스나 키보드가 동작하는 것과 마찬가지로 실행되도록 하였다. 또한, 입력 음성을 모든 reference pattern과 비교하는 것이 아니라 그 상황에 적용될 수 있는 표준 패턴을 한정함으로써 탐색시간을 줄일 수 있었으며 높은 인식률을 나타내었다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.5
• /
• pp.730-738
• /
• 1993

This paper proposes a dynamic formatting method for processing large amounts of document in a device independent manner. And it is very useful for cross-referencing among pages in a single document and for presenting multiple pages simultaneously. The method can be applied usefully to hypertext's application such as establishing a link and a cross-reference among pages in a multiple document. We implemented an electronic publishing system of WYSIWYG type using X window system and Motif graphical user interface.

• Proceedings of the KIEE Conference
• /
• 1999.07b
• /
• pp.877-879
• /
• 1999

최근 윈도우즈 기반 운영체제의 발전과 PC의 성능 향상으로 제어분야에서의 PC 활용이 가속화되면서 PC 기반의 분산제어시스템이 등장하였고, 전력연구원에서는 화력발전소의 국산개발 배연탈황설비 제어시스템에 PC 기반 분산제어시스템을 도입하여 운용하고 있다. 이러한 PC 기반 분산제어시스템의 활용으로 하위구조에 다양한 종류의 입출력시스템을 선택 가능하게 되었고 시스템 구축에 대한 독자적인 방법을 채택할 수 있게 되었다. 본 논문에서는 국산개발 배연탈황설비에 적용된 PC 기반 분산제어시스템의 전반적인 구성을 살펴보고 제어프로세서인 현장제어유닛과 하위 입출력시스템의 하드웨어 구조, 상호간 통신 구조 및 시스템 이중화구조에 대하여 기술하고자 한다. 또한 입출력시스템 간의 인터페이스 방법과 입출력 모듈 종류에 따른 현장기기의 활용방법에 대하여 알아보고 향후 PC 기반 분산제어시스템의 입출력시스템 구축 시 고려하여야 할 사항들에 대하여 기술하고자 한다.