• 한국수산과학회지
• /
• 제40권4호
• /
• pp.201-209
• /
• 2007

Line transect sampling in a sighting survey is one of most widely used methods for assessing animal abundance. This study applied distance data, collected from three sighting surveys using line transects for finless porpoise that were conducted in 2004 and 2005 off the west coast of Korea, to four models (hazard-rate, uniform, half-normal and exponential) that can use a variety of detection functions, g (x). The hazard-rate model, a derived model for the detection function, should have a shoulder condition chosen using the AIC (Akaike Information Criterion), as the most suitable model. However, it did not describe a shoulder shape for the value of g(x) near the track tine and underestimated g (x), just as the exponential model did. The hazard-rate model showed a bias toward overestimating the densities of finless porpoises with a higher coefficient of variation (CV) than the other models did. The uniform model underestimated the densities of finless porpoise but had the lowest CV. The half-normal model described a detection function with a shape similar to that of the uniform model. The half-normal model was robust for finless porpoise data and should be able to avoid density underestimation. The estimated abundance of finless porpoise was 3,602 individuals (95% CI=1,251-10,371) inshore in 2005 and 33,045 individuals (95% CI=24,274-44,985) offshore in 2004.

• 한국정보전자통신기술학회논문지
• /
• 제8권3호
• /
• pp.195-202
• /
• 2015

소프트웨어 고장분석을 위한 비동질적인 포아송과정에서 결함당 고장발생률이 상수이거나, 단조 증가 또는 단조 감소하는 패턴을 가질 수 있다. 본 논문에서는 결함의 기대값을 가정하는 유한고장소프트웨어 NHPP모형과 수리시점에서도 고장이 발생할 상황을 반영하는 무한고장 NHPP모형들을 비교 제시하였다. 소프트웨어 신뢰성분야에서 많이 사용되는 지수파우어분포에 근거한 유한고장과 무한고장 소프트웨어 신뢰성모형에 대한 비교문제를 제시하였다. 그 결과 유한고장모형이 무한고장모형보다 효율적으로 나타났다. 그리고 모수추정법은 최우추정법을 이용하였다. 이 연구를 통하여 소프트웨어 개발자들은 소프트웨어 고장현상을 파악하는데 어느 정도 도움을 줄 수 있을 것으로 사료된다.

• 방송공학회논문지
• /
• 제26권2호
• /
• pp.184-196
• /
• 2021

본 논문은 객체 검출 알고리즘을 위한 통계치 적응적인 선형 회귀 기반 객체 크기 예측 방법을 제안한다. 기존에 제안된 딥 러닝 기반 객체 검출 알고리즘 중 YOLOv2 및 YOLOv3은 객체의 크기를 예측하기 위하여 네트워크의 마지막 계층에 통계치 적응적인 지수 회귀 모델을 사용한다. 하지만, 지수 회귀 모델은 역전파 과정에서 지수 함수의 특성상 매우 큰 미분값을 네트워크의 파라미터로 전파시킬 수 있는 문제점이 있다. 따라서 본 논문에서는 미분 값의 발산 문제를 해결하기 위하여 객체 크기 예측을 위한 통계치 적응적인 선형 회귀 모델을 제안한다. 제안하는 통계치 적응적인 선형 회귀 모델은 딥러닝 네트워크의 마지막 계층에 사용되며, 학습 데이터셋에 존재하는 객체들의 크기에 대한 통계치를 이용하여 객체의 크기를 예측한다. 제안하는 방법의 성능 평가를 위하여 YOLOv3 tiny를 기반으로 제안하는 방법을 적용하여 재설계한 네트워크의 검출 성능과 YOLOv3 tiny의 검출 성능을 비교하였으며, 성능 비교를 위한 데이터셋으로는 UFPR-ALPR 데이터셋을 사용하였다. 실험을 통해 제안하는 방법의 우수성을 검증하였다.

• 품질경영학회지
• /
• 제18권2호
• /
• pp.101-116
• /
• 1990

The objectives of this thesis are : first, to estimate the parameters and Pr[X < Y] in the Marshall and Olkin's Bivariate Exponential Distribution ; and secondly, to compare the Bayes estimators of Pr[X < Y] with maximum likelihood estimator of Pr[X < Y] in the Marshall and Olkin's Bivariate Exponential Distribution. Through the Monte Carlo Simulation, we observed that the Bayes estimators of Pr[X < Y] perform better than the maximum likelihood estimator of Pr[X < Y] and the Bayes estimator of Pr[X < Y] with gamma prior distribution performs better than with vague prior distribution with respect to bias and mean squared error in the Marshall and Olkin's Bivariate Exponential Distribution.

• 대한기계학회논문집B
• /
• 제23권9호
• /
• pp.1121-1128
• /
• 1999

A single blow, transient testing method for determining the heat transfer characteristics of heat exchanger surfaces are presented. The exponential inlet temperature variations were made using screen mesh with small time constant and frontal velocities of the test section. The system is used to investigate the usefulness of a model with exponential inlet temperature variations. A data reduction program is developed to calculate the temporally and spatially averaged heat transfer coefficient using the measured disturbance and response of the fluid temperature. The results are compared with the existing theoretical and experimental data for parallel plate stacks. It was recommended to take an average for the time greater than the 99% of the final temperature had reached in order to obtain fairly good results.

• ETRI Journal
• /
• 제31권1호
• /
• pp.42-50
• /
• 2009

A new method for simulating voltage and current distributions in transmission lines is described. It gives the time domain solution of the terminal voltage and current as well as their line distributions. This is achieved by treating voltage and current distributions as distributed state variables (DSVs) and turning the transmission line equation into an ordinary differential equation. Thus the transmission line is treated like other lumped dynamic components, such as capacitors. Using backward differentiation formulae for time discretization, the DSV transmission line component is converted to a simple time domain companion model, from which its local truncation error can be derived. As the voltage and current distributions get more complicated with time, a new piecewise exponential with controllable accuracy is invented. A segmentation algorithm is also devised so that the line is dynamically bisected to guarantee that the total piecewise exponential error is a small fraction of the local truncation error. Using this approach, the user can see the line voltage and current at any point and time freely without explicitly segmenting the line before starting the simulation.

• Journal of the Korean Data and Information Science Society
• /
• 제17권2호
• /
• pp.543-557
• /
• 2006

This paper is concerned with the applicability of the chi-square approximation to the six disparity statistics: the Pearson chi-square, the generalized likelihood ratio, the power divergence, the blended weight chi-square, the blended weight Hellinger distance, and the negative exponential disparity statistic. Three dimensional contingency tables of small and moderate sample sizes are generated to be fitted to all possible hierarchical log-linear models: the completely independent model, the conditionally independent model, the partial association models, and the model with one variable independent of the other two. For models with direct solutions of expected cell counts, point estimates and confidence intervals of the 90 and 95 percentage points of six statistics are explored. For model without direct solutions, the empirical significant levels and the empirical powers of six statistics to test the significance of the three factor interaction are computed and compared.

• 한국작물학회지
• /
• 제51권4호
• /
• pp.362-368
• /
• 2006

A pot experiment was conducted to investigate the effects of soil burial depth on seedling emergences of rice (Oryza sativa) and Echinochloa spp. and to model such effects for mathematical prediction of seedling emergences. When the Gompertz curve was fitted at each soil depth, the parameter C decreased in a logistic form with increasing soil depth, while the parameter M increased in an exponential form and the parameter B appeared to be constant. The Gompertz curve was combined by incorporating the logistic model for the parameter C, the exponential model for the parameter M, and the constant for the parameter B. This combined model well described seedling emergence of rice and Echinochloa species as influenced by soil burial depth and predicted seedling emergence at a given time after sowing and a soil burial depth. Thus, the combined model can be used to simulate seedling emergence of crop sown in different soil depths and weeds present in various soil depths.

• International Journal of Computer Science & Network Security
• /
• 제21권1호
• /
• pp.192-200
• /
• 2021

In the mid of the December 2019, the virus has been started to spread from China namely Corona virus. It causes fatalities globally and WHO has been declared as pandemic in the whole world. There are different methods which can fit such types of values which obtain peak and get flattened by the time. The main aim of the paper is to find the best or nearly appropriate modeling of such data. The three different models has been deployed for the fitting of the data of Coronavirus confirmed patients in Pakistan till the date of 20^th November 2020. In this paper, we have conducted analysis based on data obtained from National Institute of Health (NIH) Islamabad and produced a forecast of COVID-19 confirmed cases as well as the number of deaths and recoveries in Pakistan using the Logistic model, Gompertz model and Auto-Regressive Integrated Moving Average Model (ARIMA) model. The fitted models revealed high exponential growth in the number of confirmed cases, deaths and recoveries in Pakistan.

• 생태와환경
• /
• 제45권4호
• /
• pp.420-443
• /
• 2012

An Individual-Based Model (IBM) was developed by employing natural and toxic survival rates of individuals to elucidate the community responses of benthic macroin-vertebrates to anthropogenic disturbance in the streams. Experimental models (dose-response and relative sensitivity) and mathematical models (power law and negative exponential distribution) were applied to determinate the individual survival rates due to acute toxicity in stressful conditions. A power law was additionally used to present the natural survival rate. Life events, covering movement, exposure to contaminants, death and reproduction, were simulated in the IBM at the individual level in small (1 m) and short (1 week) scales to produce species abundance distributions (SADs) at the community level in large (5 km) and long (1~2 years) scales. Consequently, the SADs, such as geometric series, log-series, and log-normal distribution, were accordingly observed at severely (Biological Monitoring Working Party (BMWP<10), intermediately (BMWP<40) and weakly (BMWP${\geq}50$) polluted sites. The results from a power law and negative exponential distribution were suitably fitted to the field data across the different levels of pollution, according to the Kolmogorov-Smirnov test. The IBMs incorporating natural and toxic survival rates in individuals were useful for presenting community responses to disturbances and could be utilized as an integrative tool to elucidate community establishment processes in benthic macroin-vertebrates in the streams.