• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.6
• /
• pp.14-21
• /
• 2004

This automotive headlight control system is newly proposed that, under my slope degree of the driving mad(flat up-hill, and down-hill) at night driving, the reflecting mirror of the headlight can be automatically controlled for safe driving. At first whether or not any vehicle is driven near is checked by photo sensor. Secondly, using the slope degree of the automotive feedbacked from the tilt sensor, the servo motor with the headlight is controlled to be turned right or down to the suitable angle. The servo motor is appropriately controlled according to road conditions. The proposed headlight control system is designed on the basis of the tested illumination intensity obtained according to any slope degree of roads. Finally, it is confirmed that the test model works very well in the given road conditions and environments.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.589-591
• /
• 2019

본 논문에서는 고도장비의 운용 및 정비를 위한 교육훈련 시스템 개발을 위해 자연어 처리와 딥러닝 기술을 이용하여 항공정비와 관련된 전문분야의 문서 분류가 가능한 방법을 제안하고자 한다. 문서 분류 모델의 개발을 위해 항공정비 교범을 텍스트 파일로 변환하여 총 4917개의 문서를 생성하였으며, 정비사 개인별 정비능력 관리(IMQC)를 기준으로 12개의 범주로 구분하였다. 수집된 문서는 전문분야의 문서인 점을 고려하여 전문용어 사전을 추가하였으며, KoNLPy를 이용하여 전처리를 수행하였다. 전문분야의 문서는 범주에 상관없이 문서 내용의 유사도가 매우 높은 특징을 가지고 있어, 특정 범주내에서 중요한 정도를 잘 표현 할 수 있는 TF-ICF를 이용하여 특징 추출을 하였다. 이후 합성곱 신경망(CNN)을 이용하여 특징 맵을 생성한 후 완전 결합 계층을 통하여 분류하였으며, 테스트 문서 983건을 분류한 결과 평균 73.6%의 분류성능을 보여주었다.

• Journal of Digital Convergence
• /
• v.15 no.10
• /
• pp.455-465
• /
• 2017

This study focuses on design development and verification through virtual simulation based on 3D model data in the cloud platform as a method of utilization of engineering technology of design in the fourth industrial revolution era. The goal of research is to develop and examine a design for the needs of the target that has never been met before through virtual simulations that can be conducted in practice. As a research method, we analyzed secondary data to identify the needs of the target, and did literature research for the ergonomic data and target body development stages. In addition, the design development process of this study was shown meaningful result in design, structure, safety, material, durability through loop test of 7 virtual simulations. This study can be applied to the automated process system based on 3D model data in the 4th industrial revolution era and can be used as an element of the cyber physics system for the additional research.

• Journal of Korea Multimedia Society
• /
• v.8 no.9
• /
• pp.1258-1268
• /
• 2005

While the concept of Home Network is laying by and its interests are increasing by means of digitalizing of the information communication infrastructure, many efforts are in progress toward convenient lives. Moreover, as information household appliances which have a junction of connecting to the network are appearing over the past a few years, the demands against intellectual Home Services are increasing. In this paper, by being based upon Multimedia which is an essential factor for developing of various application services on ubiquitous computing environments, we suggest a simplified application model that could apply the information to the automated processing system after studying user's behavior patterns using authentication and access control for identity certification of users. In addition, we compared captured video images in the fixed range by pixel unit through some time and checked disorder of them. And that made safe of user certification as adopting self-developed certification method which was used 'Hash' algorism through salt function of 12 byte. In order to show the usefulness of this proposed model, we did some testing by emulator for control of information after construction for Intellectual Multimedia Server, which ubiquitous network is available on as a scheme so as to check out developed applications. According to experimental results, it is very reasonable to believe that we could extend various multimedia applications in our daily lives.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.14 no.5
• /
• pp.349-361
• /
• 2021

With the development of technology and the advancement of weather forecasting models and prediction methods, higher performance weather forecasting software has been developed, and more precise and accurate weather forecasting is possible by performing software using supercomputers. In this paper, the weather forecast model used by six major countries is investigated and its characteristics are analyzed, and the Korea Meteorological Administration currently uses it in collaboration with the UK Meteorological Administration since 2012 and explains the GloSea However, the existing GloSea was conducted only on the Meteorological Administration supercomputer, making it difficult for various researchers to perform detailed research by specialized field. Therefore, this paper aims to establish a standard experimental environment in which the low-resolution version based on GloSea6 currently used in Korea can be used in local systems and test it to present the localization of low-resolution GloSea6 that can be performed in the laboratory environment. In other words, in this paper, the local portability of low-resolution Globe6 is verified by establishing a basic architecture consisting of a user terminal-calculation server-repository server and performing execution tests of the software.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.6
• /
• pp.590-596
• /
• 2020

In the island areas where system connection with the commercial power grid is difficult, it is quite important to find a method to efficiently manage energy produced with independent microgrids. In this paper, a prosumer management system for P2P power transaction was realized through the testing the power meter and the response rate of the collected data for the power produced in the small-scale microgrids in which hybrid models of solar power and wind power were implemented. The power network of the microgrid prosumer was composed of mesh structure and the P2P power transaction was tested through the power meter and DC power transmitter in the off-grid sites which were independently constructed in three places. The measurement values of the power meter showed significant results of voltage (average): 380V + 0.9V, current (average): + 0.01A, power: 1000W (-1W) with an error range within ±1%. Stabilization of the server was also confirmed with the response rate of 0.32 sec. for the main screen, 2.61 sec. for the cumulative power generation, and 0.11 sec for the power transaction through the transmission of 50 data in real time. Therefore, the proposed system was validated as a P2P power transaction system that can be used as an independent network without transmitted by Korea Electric Power Corporation (KEPCO).

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.1
• /
• pp.8-14
• /
• 2002

In this paper, we realized a real-time text-independent speaker recognition system using gaussian mixture model, and applied frame level likelihood normalization method which shows its effects in verification system. The system has three parts as front-end, training, recognition. In front-end part, cepstral mean normalization and silence removal method were applied to consider speaker's speaking variations. In training, gaussian mixture model was used for speaker's acoustic feature modeling, and maximum likelihood estimation was used for GMM parameter optimization. In recognition, likelihood score was calculated with speaker models and test data at frame level. As test sentences, we used text-independent sentences. ETRI 445 and KLE 452 database were used for training and test, and cepstrum coefficient and regressive coefficient were used as feature parameters. The experiment results show that the frame-level likelihood method's recognition result is higher than conventional method's, independently the number of registered speakers.

• Korean Journal of Remote Sensing
• /
• v.24 no.2
• /
• pp.107-115
• /
• 2008

Recently, a considerable number of studies have been conducted on the high resolution imagery showing the boundaries of objects clearly. When urban areas are analyzed in detail using the high resolution imagery, the size of analyzed zone is apt to be decided arbitrarily. Sufficient prior information about study area makes the decision of analysis zone possible; otherwise, it is difficult to determine the optimized analysis zone using only satellite imagery. In this study, the variograms of artificial simple images are analyzed before applying to the real satellite images. As a result of the analysis of simple images, the sill has an effect on the density of objects and also the size of objects and spacing influence the range. The variograms of real satellite images are analyzed with reference to the result of model test and are applied to determining the optimized analysis zone. This study shows that variogram can be applied to determining effectively the optimized analysis zone in case of no prior information on study area; moreover it will be expected to be used for an index to express the characteristics of urban imagery as well as conventional kriging and simulation.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.44 no.4 s.316
• /
• pp.36-44
• /
• 2007

In this paper, we implement the speech & face recognition system to support various ubiquitous sensor network application services such as switch control, authentication, etc. using wireless audio and image interface. The proposed system is consist of the H/W with audio and image sensor and S/W such as speech recognition algorithm using psychoacoustic model, face recognition algorithm using PCA (Principal Components Analysis) and LDPC (Low Density Parity Check). The proposed speech and face recognition systems are inserted in a HOST PC to use the sensor energy effectively. And improve the accuracy of speech and face recognition, we implement a FEC (Forward Error Correction) system Also, we optimized the simulation coefficient and test environment to effectively remove the wireless channel noises and correcting wireless channel errors. As a result, when the distance that between audio sensor and the source of voice is less then 1.5m FAR and FRR are 0.126% and 7.5% respectively. The face recognition algorithm step is limited 2 times, GAR and FAR are 98.5% and 0.036%.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.4
• /
• pp.354-358
• /
• 2001

This paper presents a fuzzy neural network model which solves the underutilization problem. This fuzzy neural network has both stability and flexibility because it uses the control structure similar to AHT(Adaptive Resonance Theory)-l neural network. And this fuzzy nenral network does not need to initialize weights and is less sensitive to noise than ART-l neural network is. The learning rule of this fuzzy neural network is the modified and fuzzified version of Kohonen learning rule and is based on the fuzzification of leaky competitive leaming and the fuzzification of conditional probability. The similarity measure of vigilance test, which is performed after selecting a winner among output neurons, is the relative distance. This relative distance considers Euclidean distance and the relative location between a datum and the prototypes of clusters. To compare the performance of the proposed fuzzy neural network with that of Kohonen Self-Organizing Feature Map the IRIS data and Gaussian-distributed data are used.