• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.43 no.1 s.343
• /
• pp.65-70
• /
• 2006

An image sensor has limited dynamic range while the human eye has logarithmic response over wide range of light intensity. Although the sensor gain can be set high to identify details in darker area on the image, this results in saturation in brighter area. The gamma correction is essential to fit the human eye response. However, the digital gamma correction degrades image quality especially for darker area on the image due to the limited ADC resolution and the dynamic range. This Paper proposes a CMOS image sensor (CIS) with a nonlinear analog-to-digital converter (AU) which performs analog gamma correction. The CIS with the proposed nonlinear analog-to-digital conversion scheme was fabricated with a $0.35{\mu}m$ CMOS process. The analog gamma correction using the proposed nonlinear ADC CIS provides the 2.2dB peak-signal-to-noise-ratio(PSM) improved image qualify than conventional digital gamma correction. The PSNR of the image obtain from the digital gamma correction is 25.6dB while it is 27.8dB for analog gamma correction. The PSNR improvement over digital gamma correction is about $28.8\%$.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.1
• /
• pp.17-27
• /
• 2014

As the importance of Cooperative Engagement Capability and network-centric warfare has been dramatically increasing, it is necessary to develop distributed tracking systems. Under the development of distributed tracking systems, it requires tracking filters and data fusion theory for nonlinear systems. Therefore, in this paper, the problem of nonlinear track fusion, which is suitable for distributed networks, is formulated, four algorithms to solve the problem of nonlinear track fusion are introduced, and performance of introduced algorithms are analyzed. It is a main problem of nonlinear track fusion that cross-covarinaces among multiple platforms are unknown. Thus, in order to solve the problem, two techniques are introduced; a simplification technique and a approximation technique. The simplification technique that help to ignore cross-covariances includes two algorithms, i.e. the sample mean algorithm and the Millman formula algorithm, and the approximation technique to obtain approximated cross-covariances utilizes two approaches, by using analytical linearization and statistical linearization based on the sigma point approach. In simulations, BCS fusion is the most efficient scheme because it reduces RMSE by approximating cross-covariances with low complexity.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.4
• /
• pp.563-568
• /
• 2007

This paper presents a methodology for predicting nonlinear time series based on the neural network with weighted fuzzy membership functions (NEWFM). The degree of classification intensity is obtained by bounded sum of weighted fuzzy membership functions extracted by NEWFM, then weighted average defuzzification is used for predicting nonlinear time series. The experimental results demonstrate that NEWFM has the classification capability of 92.22% against the target class of GDP. The time series created by NEWFM model has a relatively close approximation to the GDP which is a typical business cycle indicator, and has been proved to be a useful indicator which has the turning point forecasting capability of average 12 months in the peak point and average 6 months in the trough point during 5th to 8th cyclical period. In addition, NEWFM measures the efficiency of the economic indexes by the feature selection and enables the users to forecast with reduced numbers of 7 among 10 leading indexes while improving the classification rate from 90% to 92.22%.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.6B
• /
• pp.643-651
• /
• 2008

In the present paper, turbulent open-channel flows over longitudinal bedforms are numerically simulated. The Reynolds- averaged Navier-Stokes equations in curvilinear coordinates are solved with the non-linear $k-{\varepsilon}$ model by Speziale( 1987). First, the developed model is applied to rectangular open channel flows for purposes of model validation and parameter sensitivity studies. It is found that the parameters $C_D$ and $C_E$ are important to the intensity of secondary currents and the level of turbulent anisotropy, respectively. It is found that the non-linear $k-{\varepsilon}$ model can hardly reproduce the turbulence anisotropy near the free surface. However, the overall pattern of the secondary currents by the present model is seen to coincide with measured data. Then, numerical simulations of turbulent flows over longitudinal bedforms are performed, and the simulated results are compared with the experimental data in the literature. The simulated secondary currents clearly show upflows and downflows over the ridges and troughs, respectively. The numerical results of secondary currents, streamwise mean velocity, and turbulence structures compare favorably with the measured data. However, it is observed that the secondary currents towards the troughs were significantly weak compared with the measured data.

• Journal of Korean Society of Transportation
• /
• v.25 no.5
• /
• pp.67-77
• /
• 2007

This study deals with the traffic accidents at the unsignalized intersections in Cheongju. The purpose is to analyze the characters and the relations between road environmental factors and traffic accidents. The correlation analyses among the above factors show that the accidents are strongly related to traffic volumes and sight distances in 3-legged, and the cross angles, maximum vertical grades and sight distances in 4-legged unsignalized intersections. Also the multiple linear and nonlinear regression analyses represent that the accidents in the 3-legged increase as the traffic volume and the number of double stop-lines increase, and that the accidents in the 4-legged increase as the cross angle approaches to the 90 degree and decrease as the maximum vertical grade increases. It could be expected that this results give the good implications to the future intersection improvement projects in Cheongju.

• Nuclear Engineering and Technology
• /
• v.25 no.1
• /
• pp.110-124
• /
• 1993

A Nonlinear model-based Hybrid Controller (NHC) is developed which consists of the adaptive proportional-integral-feedforward (PIF) gains and variable structure controller. The controller has the robustness against modeling uncertainty and is applied to the trajectory tracking control of single-input, single-output nonlinear systems. The essence of the scheme is to divide the control into four different terms. Namely, the adaptive P-I-F gains and variable structure controller are used to accomplish the specific control actions by each terms. The robustness of the controller is guaranteed by the feedback of estimated uncertainty and the performance specification given by the adaptation of PIF gains using the second method of Lyapunov. The variable structure controller is incorporated to regulate the initial peak of the tracking error during the parameter adaptation is not settled yet. The newly developed NHC method is applied to the power tracking control of a nuclear reactor and the simulation results show great improvement in tracking performance compared with the conventional model-based control methods.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.4D
• /
• pp.571-576
• /
• 2011

In general, conventional road pavements are designed under the assumption that the shear strength of geomaterials are under saturated state. In reality, however, most of the pavement geomaterials exists under the unsaturated state. To deal with this gap between saturated and unsaturated conditions, in this paper, unsaturated shear strength was estimated using the results from the triaxial compression test and soil-water characteristics curves. Then, yield loads were assessed using 2-Dimensional finite element method with the selected nonlinear elastic model and the Mohr-Coulomb yield criteria. In addition, various unsaturated condition and surface layer effects on the yield load of granular materials were identified. Therefore, the results demonstrated would provide a possibility to estimate bearing capacity of paved or unpaved roads using unsaturated soil mechanics.

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.194-196
• /
• 2017

기존의 단어 기반 접근법을 이용한 개체 연결은 단어의 변형, 신조어 등이 빈번하게 나타나는 비정형 문장에 대해서는 좋은 성능을 기대하기 어렵다. 본 논문에서는 문서 임베딩과 선형 변환을 이용하여 단어 기반 접근법의 단점을 해소하는 개체 연결을 제안한다. 문서 임베딩은 하나의 문서 전체를 벡터 공간에 표현하여 문서 간 의미적 유사도를 계산할 수 있다. 본 논문에서는 또한 비교적 정형 문장인 위키백과 문장과 비정형 문장인 소셜 미디어 문장 사이에 선형 변환을 수행하여 두 문형 사이의 표현 격차를 해소하였다. 제안하는 개체 연결 방법은 대표적인 소셜 미디어인 트위터 환경 문장에서 단어 기반 접근법과 비교하여 높은 성능 향상을 보였다.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1765-1766
• /
• 2008

본 논문에서는, 전동기를 이용하는 위치제어 시스템을 이용하여, 이너루프 제어기 유무에 따라 시간 응답 특성 개선 효과에 대하여 실험을 통해 검증하고자 한다. 실험 플랜트로서, 전동기의 속도에 따라 진자의 위치를 제어하는 다중 안정도를 가지고 비선형 시스템인 아라고 진자 시스템을 이용한다. 이 시스템은 속도 정보를 알 수 있는 측정 장치가 없는데, 칼만필터로 전동기의 속도를 추정하고 속도 제어기를 설계하도록 한다. 또한, 선형 시스템을 고려하기 위하여 안정영역 동작점인 진자의 각도가 지면으로 부터 45$^{\circ}$인 선형화 모델을 이용한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.538-540
• /
• 2001

PCA는 다변수 데이터 해석법 중 가장 널리 알려진 방법 중 하나로 많은 응용을 가지고 있다. 그런데, PCA는 선형 모델이어서 비선형 구조를 분석하는데 효과적이지 않다. 이를 극복하기 위해서 PCA의 조합을 이용하는 PCA 혼합 모형이 제안되었다. PCA 혼합 모형의 핵심은 구조 선택, 즉 mixture 요소의 수와 PCA 기저의 수의 결정 인데 그의 체계적인 결정 방법이 필요하다. 본 논문에서는 단순화된 PCA 혼합 모형과 이를 위한 효율적인 구조 선택 방법을 제안한다. 각각의 mixture 요소 수에 대해서 모든 PCA 기저를 갖도록 한 상태에서 PCA 혼합 모형의 파라미터를 EM 알고리즘을 써서 결정한다. 최적의 mixture 요소의 수는 오류를 최소로 하는 것으로 결정한다. PCA 기저의 수는 PCA의 정렬성 특성을 이용해서 중요도가 적은 기저부터 하나씩 잘라 내며 오류가 최소로 하는 것으로 결정한다. 제안된 방법은 특히 다차원 데이터의 경우에 EM 학습의 횟수를 많이 줄인다. 인공 데이터에 대한 실험은 제안된 방법이 적절한 모델 구조를 결정한다는 것을 보여준다. 또, 눈 감지에 대한 실험은 제안된 방법이 실용적으로도 유용하다는 것을 보여준다.