• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.798-800
• /
• 2014

In recent years, according to the improvement of Digital image technology have been recently developed most of communication technology from multimedia communication service as well as image data transmission. But In the process of storing and transmitting noise is still generated in noise and the image degrades rapidly quality of a lot of image impulse noise. To eliminate this noise, SMF, CWMF, SWMF etc. The filters have been proposed to interfere with the noise characteristics of the filter are somewhat sufficient. Therefore, in this paper, in order to remove impulse noise is proposed a modified median filter. And impulse noise removal algorithms to confirm the existed PSNR(peak signal to noise ratio) from using conventional methods were compared.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.322-325
• /
• 2018

영상의 관심 영역 검출은 영상처리 및 컴퓨터 비전 응용 분야에서 꾸준하게 사용되고 있는 기법이다. 특히, 근래 심층신경망 연구의 급격한 발전에 힘입어 심층신경망을 이용한 관심 영역 검출 기법에 대한 연구가 활발하게 진행되고 있다. 한편 Fully Convolutional Network(이하 FCN)은 본래 심층 예측(Dense Prediction)을 통한 의미론적 영상 분할(Semantic Segmentation)을 수행하기 위해 제안된 심층신경망 구조이다. FCN을 영상의 관심 영역 검출에 활용하여도 기존 관심 영역 검출 기법과 비교하여 충분히 좋은 성능을 발휘할 수 있다. 그러나 FCN에 사용되는 convolution 층의 수가 많고, 이에 따른 가중치(weight)의 개수도 기하급수적으로 늘어나 검출에 필요한 시간 복잡도가 매우 크다는 문제점이 있다. 따라서 본 논문에서는 기존 FCN이 가진 검출 시간 복잡도의 문제점을 convolution 층의 가중치 관점에서 해결하고자 이를 조절하여 FCN의 관심 영역 검출 속도를 향상시키는 방법을 제안한다. 적절한 convolution 층의 가중치를 조절함으로써, MSRA10K 데이터셋 환경에서 검출 정확도를 크게 저하시키지 않고도 최대 약 20.5%만큼 검출 속도를 향상시킬 수 있었다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.133-135
• /
• 2022

With the development of IoT and AI technologies and media, various digital devices are being used, and unmanned and automation is progressing rapidly. In particular, high-level image processing technology is required in fields such as smart factories, autonomous driving technology, and intelligent CCTV. However, noise present in the image affects processes such as edge detection and object recognition, and causes deterioration of system accuracy and reliability. In this paper, we propose a filtering algorithm using fuzzy weights to reconstruct images damaged by complex noise. The proposed algorithm obtains a reference value using noise judgment and calculates the final output by applying a fuzzy weight. Simulation was conducted to verify the performance of the proposed algorithm, and the result image was compared with the existing filter algorithm and evaluated.

• Journal of Energy Engineering
• /
• v.12 no.1
• /
• pp.1-10
• /
• 2003

Pressurized fluidized bed combustion unit is operated at pressures of 1~1.5 MPa with combustion temperatures of 850~87$0^{\circ}C$. The pressurized coal combustion system heats steam, in conventional heat transfer tubing, and produces a hot gas supplied to a gas turbine. Gas cleaning is a vital aspect of the system, as is the ability of the turbine to cope with some residual solids. The need to pressurize the feed coal, limestone and combustion air, and to depressurize the flue gases and the ash removal system introduces some significant operating complications. The proportion of power coming from the steam : gas turbines is approximately 80:20%. Pressurized fluidized bed combustion and generation by the combined cycle route involves unique control considerations, as the combustor and gas turbine have to be properly matched through the whole operating range. The gas turbines are rather special, in that the maximum gas temperature available from the FBC is limited by ash fusion characteristics. As no ash softening should take place, the maximum gas temperature is around 90$0^{\circ}C$. As a result a high pressure ratio gas turbine with compression intercooling is used. This is to offset the effects of the relatively low temperature at the turbine inlet.

• 한국신재생에너지학회:학술대회논문집
• /
• 2009.06a
• /
• pp.880-883
• /
• 2009

석탄가스화복합발전(IGCC) 시스템에서 합성가스 중에 포함된 황화수소($H_2S$)는 후단의 가스 터빈과 같은 장치의 부식을 방지하고, 합성가스를 이용하는 연료전지 등의 연계 공정에서 요구하는 수준에 맞추어 정제되어야 한다. 본 연구에서는 $H_2S$ 정제공정 추가에 따른 IGCC 시스템의 효율저하를 최소화하기 위하여 고온고압에서 사용가능한 탈황제를 분무건조법을 이용하여 제조하고 제조된 탈황제에 대해 물성 및 황 흡수능 시험을 실시하였다. 형상, 내마모도, 평균입자크기, 충진밀도와 함께 제조된 탈황제가 적용되는 유동층 공정에 적합한 강도를 보유하는지 여부를 미국표준시험방법에 의하여 측정하였다. 황 흡수능은 열중량분석기를 반응기로 사용하여 모사 합성 가스 분위기에서 측정하였다. 분무건조 성형된 탈황제의 일부가 구형이 아닌 타원형 또는 도넛 형태를 나타내고 있어 형상 개선을 위한 제조방법 개선이 필요한 것으로 나타났다. 제조된 탈황제는 기공도가 65% 이상으로 macropore가 기공부피의 대부분을, mesopore가 비표면적의 대부분을 제공하고 있었다. 소성온도를 650 $^{\circ}C$에서 750 $^{\circ}C$로 증가시킴에 따라 대체로 강도가 감소하는 경향을 나타내었다. 열중량분석기로 측정된 황 흡수능은 약 10 wt%로 나타났다. 제조된 탈황제 중 일부는 유동층 공정에 적합한 물성을 보유하고 있었으며 반응성 또한 기존에 개발된 탈황제에 버금가는 성능을 나타내어 향후 공정 적용이 가능할 것으로 분석되었다.

• Proceedings of the KSR Conference
• /
• 2010.06a
• /
• pp.204-212
• /
• 2010

Currents which are generated at both ends of drive motor by operating brake pedal when subway train is driving, will generate the reviltalization actual effect if they are not used immediately. So there exist rolling stock established Dynamic braking annex for the purpose of stable brake performance in case there are no retrogress train around or no stable wiring voltage.Therefore 55% of entered energy are consumed in subway train. 45% are used in down gradient section or for regeneration energy and among them 25% are used for another train's retrogression through the wiring. So to reduce reviltalization method keeping the existing system, substation's service voltage should be declined about 5%. And then it will ease off excessive wiring voltage rise. And there need energy reduction by flexible service voltage adjustment and study for energy consumption efficiency in the subway.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.1
• /
• pp.157-167
• /
• 2020

The big data and artificial intelligence technology, which has provided the foundation for the recent 4th industrial revolution, has become a major driving force in business innovation across industries. In the field of information security, we are trying to develop and improve an intelligent security system by applying these techniques to large-scale log data, which has been difficult to find effective utilization methods before. The quality of security log big data, which is the basis of information security AI learning, is an important input factor that determines the performance of intelligent security system. However, the difference and complexity of log data by various product has a problem that requires excessive time and effort in preprocessing big data with poor data quality. In this study, we research and analyze the cases related to log data collection of various firewall. By proposing firewall log data collection format standard, we hope to contribute to the development of intelligent security systems based on security log big data.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.6
• /
• pp.1235-1242
• /
• 2018

When the wind speed changes rapidly, the wind turbine is stopped for the safety of the power system and the mechanical system. At that moment, the wind turbine gearbox is damaged and broken due to the contact load of the gearbox. In addition, the problems such as increasing frictional heat and deteriorate of the brake occur, because the power of the blades is transmitted directly to the brakes. This paper proposes a hybrid magnetic gear shape that solves the problem caused by the contact of the mechanical gear, which is the power transmission device of the wind power generation system, and the power cutoff system. The shape of the hybrid magnetic gearsuitable for the wind power generation system is derived through the torque and loss analysis according to the shape of the hybrid magnetic gear by using the two dimensional finite analysis method.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.1
• /
• pp.87-92
• /
• 2023

With the continuous development of the speech recognition system, the recognition rate for speech has developed rapidly, but it has a disadvantage in that it cannot accurately recognize the voice due to the noise generated by mixing various voices with the noise in the use environment. In order to increase the vocabulary recognition rate when processing speech with environmental noise, noise must be removed. Even in the existing HMM, CHMM, GMM, and DNN applied with AI models, unexpected noise occurs or quantization noise is basically added to the digital signal. When this happens, the source signal is altered or corrupted, which lowers the recognition rate. To solve this problem, each voice In order to efficiently extract the features of the speech signal for the frame, the MFCC was improved and processed. To remove the noise from the speech signal, the noise removal method using the Gaussian model applied noise deviation estimation was improved and applied. The performance evaluation of the proposed model was processed using a cross-correlation coefficient to evaluate the accuracy of speech. As a result of evaluating the recognition rate of the proposed method, it was confirmed that the difference in the average value of the correlation coefficient was improved by 0.53 dB.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.221-229
• /
• 2023

Recently, Grid-forming(GFM) converter, which offers features such as virtual inertia, damping, black start capability, and islanded mode operation in power systems, has gained significant attention. However, in low-voltage microgrids(MG), it faces challenges due to the coupling phenomenon between active and reactive power caused by the low line impedance X/R ratio and a non-negligible power angle. This power coupling issue leads to stability and performance degradation, inaccurate power sharing, and control parameter design problems for GFM converters. Therefore, this paper serves as a review study on not only control methods associated with GFM converters but also power decoupling techniques. The aim is to introduce promising control methods and enhance accessibility to future research activities by providing a critical review of power decoupling methods. Consequently, by facilitating easy access for future researchers to the study of power decoupling methods, this work is expected to contribute to the expansion of distributed power generation.