• The Transactions of the Korea Information Processing Society
• /
• v.5 no.9
• /
• pp.2294-2301
• /
• 1998

The aim of this paper is to extract features from each news scenes for example, symbol icon which can be distinct each broadcasting corp, icon and caption which are has feature and important information for the scene in respectively, In this paper, we propose extraction methods of caption that has important prohlem of news videos and it can be classified in three steps, First of al!, we converted that input images from video frame to YIQ color vector in first stage. And then, we divide input image into regions in clear hy using equalized color histogram of input image, In last, we extracts caption using edge histogram based on vertical and horizontal line, We also propose the method which can extract news icon in selected key frames by the difference of inter-histogram and can divide each scene by the extracted icon. In this paper, we used comparison method of edge histogram instead of complex methcxls based on color histogram or wavelet or moving objects, so we shorten computation through using simpler algorithm. and we shown good result of feature's extraction.

• Proceedings of the KIEE Conference
• /
• 2006.07d
• /
• pp.2134-2135
• /
• 2006

전력거래소는 발전경쟁시장(CBP; Cost-Based Pool) 장기화에 따른 운영상의 효율성을 개선하고 기 개발된 시장운영시스템(MOS; Market Operation System)을 활용하여 급전체계를 개선하기 위해 준비중이다. 현행 급전체계에서는 거래 전일에 수행한 수요예측을 바탕으로 1시간 단위운영발전계획을 전일에 수립하고 EMS(Energy Management System)를 이용하여 발전기에 대한 경제부하배분(ED; Economic Dispatch)을 시행하고 있지만, 현 EMS는 시장체제 환경 전에 도입된 설비로 시장환경에 대한 고려가 되어 있지 않고 계통운영 보조서비스의 실시간 반영이 어려운 점이 있다. 전력거래소는 실시간 급전 운영을 위해 기존 EMS에 MOS를 연계하여 MOS의 5분 단위 수요예측량을 기반으로 송전망 제약과 예비력 요구량 등을 고려한 발전기별 경제부하 배분량 및 예비력 배분량을 결정하고, 추가적으로 EMS에서 수요예측 오차 및 주파수 보정량을 실시간으로 계산하여 발전기별로 배분하도록 함으로써, 1일 전 시행하던 급전계획을 취득 자료를 기반으로 5분 단위로 실시간 계산할 수 있도록 급전 체계를 개선할 계획이다. 이를 통해 실시간으로 에너지와 예비력을 동시에 최적화함으로 전력시장 및 전력계통 운영을 한층 선진화 할 수 있는 계기를 마련하였으며 또한 저비용 발전기 사용을 극대화함으로 발전비용의 절감에도 기여하는 효과를 기대할 수 있다. MOS-EMS간 자료연계에는 ICCP(Inter-Control Communication Protocol)와 FTP 프로토콜을 사용하였고, 수차례 모의운영을 통하여 데이터베이스 및 현장 취득 자료의 정확도(accuracy)가 양 시스템 간 연계 및 전력 계통의 안정적 운영에 매우 중요한 요소로 나타났다. 전력거래소는 장기적으로 CIM(Common Information Model)기반의 표준 전력계통 데이터베이스를 구축하고 시스템 간 자료 연계를 위해 XML을 활용하여 시스템 간 상호 운영성(Interoperability)과 자료 연계의 안정성을 높일 계획이다. 본 논문은 MOS-EMS 연계에 따른 시스템 간 자료의 흐름 및 처리에 대해 주로 설명하고자 한다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.5
• /
• pp.23-34
• /
• 2004

In this paper we propose a feature point tracking algorithm using optical flow under non-prior taming active feature model (NPT-AFM). The proposed algorithm mainly focuses on analysis non-rigid objects[1], and provides real-time, robust tracking by NPT-AFM. NPT-AFM algorithm can be divided into two steps: (i) localization of an object-of-interest and (ii) prediction and correction of the object position by utilizing the inter-frame information. The localization step was realized by using a modified Shi-Tomasi's feature tracking algoriam[2] after motion-based segmentation. In the prediction-correction step, given feature points are continuously tracked by using optical flow method[3] and if a feature point cannot be properly tracked, temporal and spatial prediction schemes can be employed for that point until it becomes uncovered again. Feature points inside an object are estimated instead of its shape boundary, and are updated an element of the training set for AFH Experimental results, show that the proposed NPT-AFM-based algerian can robustly track non-rigid objects in real-time.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.316-316
• /
• 2016

HyGIS는 DEM 기반의 수문지형처리를 중심으로 다양한 모형을 연계할 수 있도록 구성된 S/W 모음이다. 이는 한국건설기술연구원을 비롯한 다수의 기관 및 연구원들이 노력한 성과물이다. 본 연구는 기존 HyGIS 연구과정에서 도출된 성과물의 실용화 사업화를 위한 방안을 연구하였다. 이를 위하여 S/W 테스팅, 오픈소스 도입, 클라우드 컴퓨팅으로 나누어 접근하였다. 먼저 S/W의 테스팅에 있어서 기존 개발 소스코드는 블랙박스 테스트 방식의 동등 클래스 분할, 경계 값 분석 등 일부 모듈에 대한 단위 테스트와 제한적인 통합테스트가 수행된 바 있다. 보다 체계적인 테스트 단계로서 화이트박스 테스트 개념 중 문장/분기/조건 커버리지에 대하여 검토하였으며, 실제 소스코드 중 핵심 구간에 대한 적용 및 정량화를 통하여 현 수준을 객관적으로 진단하였고 보완 방안을 도출하였다. 오픈소스 적용을 위하여 QGIS, MapWindow 등 공간정보 분야의 최신 오픈소스 모듈을 비교 검토하였다. 적용 단계는 이를 기존 HyGIS S/W에 반영시키는 과정이며, S/W 관점에서는 컴포넌트 모듈의 대체라고 표현될 수 있다. 대규모의 전환 비용이 발생되므로 적용 후보에 대하여는 기능적 측면 뿐만 아니라 마이그레이션 비용과 중장기적인 유지보수 비용을 고려한 검토가 이루어 졌다. 한편 오픈소스 기술의 적용은 단순히 구성 요소 원가절감 측면만이 아닌, 중장기적 유지보수 체계 도모 및 지속가능한 생태계로의 전환에 더 큰 의의가 있다. 마지막으로 클라우드 컴퓨팅 기술의 적용 분야이다. HyGIS 입력 Data의 공급을 위한 인프라로서 자체 구축 인프라가 아닌 IaaS 클라우드인 Blob Storage 및 CDN을 시험 적용하였다. 클라우드를 활용함으로써 초기 비용을 최소화하고 합리적 비용으로 유연한 확장이 가능한(Scale Out, Scale Up) 구조를 취하게 되었다. 또한 입력 Data 공급 서버를 위한 Storage 측면만이 아니라 S/W의 배포에 있어서도 클라우드 컴퓨팅 기술을 활용하고자 시도하였다. 클라우드 기술을 활용하여 HyGIS S/W가 설치된 VM(Virtual Machine)자체를 임대하는 방식으로 시험 구성 되었다. VM에 대한 RDP 프로토콜 Access에 있어서 IP기반 접근 제어를 통하여 보안을 강화하는 방안을 실험하였으며, ISO 27001, ISO 27018 등 관련 보안 규정에 부합하는 서비스 제공이 가능하도록 검토하였다. 이러한 클라우드 VM방식 서비스를 통하여 Package형 S/W 뿐만 아니라 Subscription 방식의 서비스 제공 방식을 병행할 수 있다. 사용자에게는 S/W 설치 및 H/W Lock 구비 과정이 생략되는 이점이 있다.

• Journal of Information Management
• /
• v.42 no.1
• /
• pp.85-111
• /
• 2011

The purpose of this study was to seek a workable alternative to replace a big deal related to the journal budget for the maintenance of academic libraries with the largest issue on the E-journal consortium. The contents of this study was to present it. It had examined the current situation, strengths, weaknesses and corresponding to replace the big deal contract. After reviewing the literature, we looked into the alternative activities for the big deal such as open access-based, usage-based, consortium improvement-based, publishers lead, and other models. As a result, the 'consortium cost reapportion model' was an alternative for the KESLI. The alternative was in the short term for cost division format, but long-term oriented for a consortium single(bloc) payment type or national licence model. The model was based on the data from the last year. It had evaluated download the PDF and HTML documents, but the three times weighting more than others, and the rest of 14 factors of 0.5 to 5 out of 100 total score. The total amount negotiated by national units 10, 20 and 30 grades for the final step was allocated to the participating library on the KESLI consortium.

• Journal of Korea Multimedia Society
• /
• v.16 no.12
• /
• pp.1357-1367
• /
• 2013

Recently, as one of images based methods in facial expression recognition, the research which used ULBP block histogram feature and SVM classifier was performed. Due to the properties of LBP introduced by Ojala, such as highly distinction capability, durability to the illumination changes and simple operation, LBP is widely used in the field of image recognition. In this paper, we combined $LBP_{8,2}$ and $LBP_{8,1}$ to describe micro features in addition to shift, size change in calculating ULBP block histogram. From sub-windows of 660 of $LBP_{8,1}$ and 550 of $LBP_{8,2}$, ULBP histogram feature of 1210 were extracted and weak classifiers of 50 were generated using AdaBoost. By using the combined $LBP_{8,1}$ and $LBP_{8,2}$ hybrid type of ULBP histogram feature and SVM classifier, facial expression recognition rate could be improved and it was confirmed through various experiments. Facial expression recognition rate of 96.3% by hybrid boosted ULBP block histogram showed the superiority of the proposed method.

• The Journal of the Korea Contents Association
• /
• v.22 no.5
• /
• pp.288-304
• /
• 2022

Walking is an eco-friendly and sustainable means of transportation that promotes health and endurance. Despite the positive health benefits of walking, pedestrian safety is a serious problem in Korea. Therefore, it is necessary to investigate with various studies to reduce pedestrian-vehicle crashes. In this study, the cognitive characteristics affecting pedestrian-vehicle crashes were considered by applying deep learning semantic segmentation. The main results are as follows. First, it was found that the risk of pedestrian-vehicle crashes increased when the ratio of buildings among cognitive factors increased and when the ratio of vegetation and the ratio of sky decreased. Second, the humps were shown to reduce the risk of pedestrian-related collisions. Third, the risk of pedestrian-vehicle crashes was found to increase in areas with many neighborhood roads with lower hierarchy. Fourth, traffic lights, crosswalks, and traffic signs do not have a practical effect on reducing pedestrian-vehicle crashes. This study considered existing physical neighborhood environmental factors as well as factors in cognitive aspects that comprise the visual elements of the streetscape. In fact, the cognitive characteristics were shown to have an effect on the occurrence of pedestrian- related collisions. Therefore, it is expected that this study will be used as fundamental research to create a pedestrian-friendly urban environment considering cognitive characteristics in the future.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.5
• /
• pp.341-346
• /
• 2013

In this paper, we propose a rule extraction method using a modified Fuzzy Min-Max (FMM) neural network. The suggested method supplements the hyperbox definition with a frequency factor of feature values in the learning data set. We have defined a relevance factor between features and pattern classes. The proposed model can solve the ambiguity problem without using the overlapping test process and the contraction process. The hyperbox membership function based on the fuzzy partitions is defined for each dimension of a pattern class. The weight values are trained by the feature range and the frequency of feature values. The excitatory features and the inhibitory features can be classified by the proposed method and they can be used for the rule generation process. From the experiments of sign language recognition, the proposed method is evaluated empirically.

• Journal of Broadcast Engineering
• /
• v.11 no.3 s.32
• /
• pp.326-335
• /
• 2006

Design criteria far a block code which guarantees minimized PAPR of the OFDM signals are proposed in this paper. Encoding procedure of the minimum PAPR codes (MPC) is composed of searching a seed codeword, circular shifting the register elements, and determining codeword inversion. It is shown that the PEP is invariant to the circular shift of register elements and codeword inversion. Based on such properties, systematic encoding rule for MPC is proposed. In addition proposed encoding rule can reduced greatly the size of look up table for MPC.

• Journal of Digital Contents Society
• /
• v.10 no.4
• /
• pp.625-632
• /
• 2009

In these days mobile handsets have come to be used at almost every user. The performance improvement of mobile devices and networks have made this trend possible. As a great variety of mobile applications are published, the necessity of running large-scale mobile applications becomes greater than before. To accomplish this, the existing researchers have developed mobile cluster computing libraries like Mobile-JPVM. In this paper, we implement a compute-intensive Animated GIF generating application and its cell phone viewer software using Mobile-JPVM library. We find out by the real execution of our softwares on the KTF handsets that they can sufficiently run on cellular phones. Our Animated GIF generator and its viewer are going to be commercially used for the mobile fashion advertisement systems.