• ETRI Journal
• /
• 제45권1호
• /
• pp.119-130
• /
• 2023

We propose a quantized gradient search algorithm that can achieve global optimization by monotonically reducing the quantization step with respect to time when quantization is composed of integer or fixed-point fractional values applied to an optimization algorithm. According to the white noise hypothesis states, a quantization step is sufficiently small and the quantization is well defined, the round-off error caused by quantization can be regarded as a random variable with identically independent distribution. Thus, we rewrite the searching equation based on a gradient descent as a stochastic differential equation and obtain the monotonically decreasing rate of the quantization step, enabling the global optimization by stochastic analysis for deriving an objective function. Consequently, when the search equation is quantized by a monotonically decreasing quantization step, which suitably reduces the round-off error, we can derive the searching algorithm evolving from an optimization algorithm. Numerical simulations indicate that due to the property of quantization-based global optimization, the proposed algorithm shows better optimization performance on a search space to each iteration than the conventional algorithm with a higher success rate and fewer iterations.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.408-412
• /
• 2009

In this paper, a new efficient algorithm for global motion estimation is proposed. This algorithm uses a previous 4-parameter model based global motion estimation algorithm and M-estimator for improving the accuracy and robustness of the estimate. The first algorithm uses the block based motion vector fields and which generates a coarse global motion parameters. And second algorithm is M-estimator technique for getting precise global motion parameters. This technique does not increase the computational complexity significantly, while providing good results in terms of estimation accuracy. In this work, an initial estimation for the global motion parameters is obtained using simple 4-parameter global motion estimation approach. The parameters are then refined using M-estimator technique. This combined algorithm shows significant reduction in mean compensation error and shows performance improvement over simple 4-parameter global motion estimation approach.

• Journal of applied mathematics & informatics
• /
• 제24권1_2호
• /
• pp.245-259
• /
• 2007

This paper studies some properties of the value functions and gives some sufficient and necessary conditions about the presented global error and local error. And it leads to one kind of relationship between iterative points and optimal solution or K-T point.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.15-17
• /
• 2006

This paper presents a new global intensity compensation method for extracting moving object in a visual surveillance system by compensating time variant intensity changes of background region. The method that compensates a little changes of intensity due to time variant illumination change and automatic gain control of camera is called global intensity compensation. The proposed method expresses global intensity change with a mapping table to describe complex form of intensity change while the previous method models this global intensity change with a simple function as a straight line. The proposed method builds the mapping table by calculating the cross histogram between two images and then by selecting an initial point for generating the mapping table by using Hough transform applied to the cross histogram image. Then starting from the initial point, the mapping table is generated according to the proposed algorithm based on the assumption that reflects the characteristic of global intensity change. Experimental results show that the proposed method makes the compensation error much smaller than the previous GIC method

• 벤처창업연구
• /
• 제16권3호
• /
• pp.73-86
• /
• 2021

이 연구의 목적은 기업가적 혁신성을 정확하게 예측하는 고도화된 분석 모델을 탐색하는 것이다. 기업가정신 연구 분야에서는 최초로, 데이터 과학적 접근방식에 해당되는 기계학습(Machine learning)을 이용해 기업가적 혁신성(entrepreneurial innovativeness)을 예측하는 모델을 제시한다. 예측모델을 구축하기 위하여 Global Entrepreneurship Monitor(GEM)의 62개국 22,099건 데이터를 이용한다. 27개 설명변수로 이뤄진 데이터 셋을 토대로 전통적 통계방법인 다중회귀분석과, 회귀트리, 랜덤포레스트, XG부스트, 인공신경망 등 기계학습을 이용한 예측모델을 구축하고 각 모델의 성능을 비교한다. 모델의 성능 평가를 위해 RMSE(Root mean square error), MAE(Mean absolute error)와 상관관계(Correlation) 등 지표를 사용한다. 분석 결과 5가지 기계학습 기반 모델은 모두 전통적 방법에 비해 우수한 성능을 보였으며, 예측 성능이 가장 좋은 모델은 XG부스트였다. XG부스트를 통한 기업가적 혁신성 예측에 있어서 기여도가 높은 변수는 창업가의 기회인지 및 시장 확장의 교차항 변수이며, 이는 신시장에서 기회를 획득하고자 하는 유형의 창업기업이 높은 혁신성을 보인다는 점을 확인했다. 이 연구는 고도화된 분석방법인 기계학습을 이용해 새로운 예측모델을 제시, 기업가정신 연구의 시야를 확장했다는 점에서 의의를 지닌다.

• 대기
• /
• 제17권1호
• /
• pp.1-15
• /
• 2007

A constraint based on climatology is introduced to the cost function of the three-dimensional variational assimilation (3dVar) to correct the error of the zonal mean wind structure in the global data assimilation system at Korea Meteorological Administration (KMA). The revised cost function compels the analysis fit to the chosen climatology while keeping the balance between the variables in the course of analysis. The constraint varies selectively with the vertical level and the horizontal scale of the motion. The zonally averaged wind field from European Centre for Medium-Range Weather Forecasts Re-Analysis 40 (ERA-40) is used as a climatology field in the constraint. The constraint controls only the zonally averaged stratospheric long waves with total wave number less than 20 to fix the error of the large scale wind field in the stratosphere. The constrained 3dVar successfully suppresses the erroneous westerly in the stratospheric analysis promptly, and has been applied on the operational global 3dVar system at KMA.

• Journal of Astronomy and Space Sciences
• /
• 제27권2호
• /
• pp.117-122
• /
• 2010

An ionospheric error simulation is needed for creating precise Global Positioning System (GPS) signal using GPS simulator. In this paper we developed Klobuchar coefficients n ${\alpha}_n$ and ${\beta}_n$ (n = 1, 2, 3, 4) generation algorithms for simulator and verified accuracy of the algorithm. The algorithm extract those Klobuchar coefficients from broadcast (BRDC) messages provided by International GNSS Service during three years from 2006 through 2008 and curve-fit them with sinusoidal and linear functions or constant. The generated coefficients from our developed algorithms are referred to as MODL coefficients, while those coefficients from BRDC messages are named as BRDC coefficients. The maximum correlation coefficient between MODL and BRDC coefficients was found for ${\alpha}_2$ and the value was 0.94. On the other hand, the minimum correlation was 0.64 for the case of ${\alpha}_1$. We estimated vertical total electron content using the Klobuchar model with MODL coefficients, and compared the result with those from the BRDC model and global ionosphere maps. As a result, the maximum RMS was 3.92 and 7.90 TECU, respectively.

• 한국수자원학회논문집
• /
• 제52권1호
• /
• pp.11-19
• /
• 2019

본 연구에서는 인공위성 및 재분석 자료인 Global Land Data Assimilation System (GLDAS), Global Land Evaporation Amsterdam Model (GLEAM), MOD16의 실제증발산량 산출물을 활용하여 한국수자원조사기술원(Korea Institute of Hydrological Survey, KIHS)에서 관리하고 있는 청미천(cheongmicheon farmland site, CFK)과 설마천(seolmacheon site, SMK) flux tower에서 검증하였고, Triple collocation (TC) 방법을 활용하여 자료간의 불확실성 및 상관성분석을 수행하였다. 플럭스타워와의 검증 결과에서는 전반적으로 GLEAM>GLDAS>MOD16순으로 좋은 결과를 나타내었으며, 세가지 산출물의 조합(S1: flux tower vs. GLDAS vs. MOD16, S2: flux tower vs. GLDAS vs. GLEAM, S3: flux tower vs. GLEAM vs. MOD16)을 통한 TC 결과에서는 청미천(설마천)에서 GLEAM>GLDAS>MOD16>flux tower (GLDAS>GLEAM>MOD16>flux tower)순으로 좋은 결과를 나타내었다. TC 분석 결과에서 Flux tower의 error variance와 correlation coefficient가 상대적으로 좋은 결과를 나타내지 못하였으므로, 한반도 지역에서 인공위성과 재분석 자료(GLDAS vs. GLEAM vs. MOD16)만을 활용하여 TC를 적용하였다. 그 결과, GLDAS와 GLEAM이 한반도 영역에서 낮은 error variance 와 높은 correlation coefficient를 나타낸 반면, MOD16의 경우, 농지에서 낮은 correlation coefficient과 높은 error variance를 나타내었다.

• Kyungpook Mathematical Journal
• /
• 제56권2호
• /
• pp.311-327
• /
• 2016

In this paper, we propose an error embedded Runge-Kutta method to improve the traditional embedded Runge-Kutta method. The proposed scheme can be applied into most explicit embedded Runge-Kutta methods. At each integration step, the proposed method is comprised of two equations for the solution and the error, respectively. These solution and error are obtained by solving an initial value problem whose solution has the information of the error at each integration step. The constructed algorithm controls both the error and the time step size simultaneously and possesses a good performance in the computational cost compared to the original method. For the assessment of the effectiveness, the van der Pol equation and another one having a difficulty for the global error control are numerically solved. Finally, a two-body Kepler problem is also used to assess the efficiency of the proposed algorithm.

• 한국항행학회논문지
• /
• 제14권5호
• /
• pp.648-654
• /
• 2010

본 논문에서는 위성항법 기반의 위치 정보를 이용하여 주행하는 AGV(Autonomous Guided Vehicle)의 안전성을 향상시키기 위한 항법 시스템을 구성하고 성능 시험을 수행하였다. 이를 위해 위성항법 신호에 급격한 오차를 감지하고 위성항법 신호가 단절된 경우에도 연속적인 주행이 가능하도록 DR(Dead Reckoning) 항법 시스템을 구성하였다. 주행 시험 결과 0.15m이상의 위성항법 오차를 감지할 수 있었으며 8초의 위성항법 신호 단절에서 약 1.5m 이내의 오차로 안정적인 주행을 확인할 수 있었다.