• Journal of the Korean Data and Information Science Society
• /
• v.21 no.2
• /
• pp.241-250
• /
• 2010

Over the past few decades, several studies have been made on the modeling of circular data. But these studies focused mainly on the symmetrical cases including von Mises distribution. Recently, many studies with skew-normal distribution have been conducted in the linear case. In this paper, we dealt the problem of fitting of non-symmetrical circular data with wrapped skew-normal distribution which can be derived by using the principle of wrapping. Wrapped skew-normal distribution is very flexible to asymmetical data as well as to symmetrical data. Multi-modal data are also fitted by using the mixture of wrapped skew-normal distributions. To estimate the parameters of mixture, we suggested the EM algorithm. Finally we verified the accuracy of the suggested algorithm through simulation studies. Application with real data is also considered.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.4
• /
• pp.1209-1217
• /
• 2006

For the mixture of shifted Poisson distributions, a method of parameter estimation is proposed. The range of the shifted parameters are estimated first and for each shifted parameter set EM algorithm is applied to estimate the other parameters of the distribution. Among the estimated parameter sets, one with minimum likelihood for given data is to be set as the final estimate. In simulation experiments, the suggested estimation method shows to have a good performance.

• Journal of Broadcast Engineering
• /
• v.19 no.5
• /
• pp.656-664
• /
• 2014

In the case of a haze image, transferring the information of the original image is difficult as the contrast leans toward bright regions. Thus, dehazing algorithms have become an important area of study. Normally, since it is hard to obtain a haze-free image, the output image is qualitatively analyzed to test the performance of an algorithm. However, this paper proposes a quantitative error comparison based on reproducing the haze image using a disparity map. In addition, a Hidden Random Markov Model and EM algorithm are used to remove any block artifacts. The performance of the proposed algorithm is confirmed using a variety of synthetic and natural images.

• Journal of Internet of Things and Convergence
• /
• v.10 no.3
• /
• pp.25-33
• /
• 2024

In tomographic image reconstruction, the focus is on developing CT image reconstruction methods that can maintain high image quality while reducing patient radiation exposure. Typically, statistical image reconstruction methods have the ability to generate high-quality and accurate images while significantly reducing patient radiation exposure. However, in cases like CT image reconstruction, which involve multi-dimensional parameter estimation, the degree of the Hessian matrix of the penalty function is very large, making it impossible to calculate. To solve this problem, the author proposed the PEMG-1 algorithm. However, the PEMG-1 algorithm has issues with the convergence speed, which is typical of statistical image reconstruction methods, and increasing the penalty log-likelihood. In this study, we propose a reconstruction algorithm that ensures fast convergence speed and monotonic increase in likelihood. The basic structure of this algorithm involves sequentially updating groups of pixels instead of updating all parameters simultaneously with each iteration.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.27 no.5
• /
• pp.449-458
• /
• 2007

In this paper, we reconstructed the cross-sectional images of two phantoms simulating a petrochemical process from gamma radiation measurements. Three different weight functions for EM image reconstruction algorithm were built and compared with histograms representing the variance of the homogeneity of the phantom material, The radiation source, $^{137}Cs$, collimated by a lead with 5 mm diameter aperture and the measurement was made with a lead shielded 1inch NaI detector. As a result, the method taking into account the beam area in each pixel for a weight function showed the best resolution among the three methods.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.283-284
• /
• 2021

본 논문에서는 서로 다른 진화 전략의 병렬화를 구현하기 위해 섬 모델을 도입하고 자원 간의 균형을 유지하기 위해 Monod 모델을 활용하는 PDE-EM이라는 생태 모델 알고리즘을 기반으로 한 새로운 병렬 DE를 제안하도록 한다. 각 섬은 동일한 자원으로 서로 다른 전략으로 진화한다. 지정된 세대 수마다 섬의 진화 정도에 따라 등급이 매겨지고, Monod 모델을 활용하여 각 섬에 다양한 자원이 할당된다.

• The Korean Journal of Applied Statistics
• /
• v.22 no.2
• /
• pp.365-373
• /
• 2009

Since 1990, identifying the sex of fetus and illegal abortion has brought the sex ratio imbalance at birth in Korea due to a notion of preferring a son to a daughter, socio-economic development, population policy, and so forth. Although there have been many researches such as time series analysis and region difference analysis to monitor this sex ratio imbalance, they have a defect that time and space could not be included in the analysis simultaneously. This study analyzes the sex ratio imbalance at birth, taking into account time and region at the same time. The analysis considered the numbers of male and female babies, who were born as the third or latter in their families, in 2000 and 2001 at 234 Gu / Si / Goon administrative districts. Here, we suggest a mixture model of binomial distributions, assuming heterogeneous populations. The estimation of the location parameters, weights and correlation coefficient of the mixture model is conducted by the EM algorithm, and the heterogeneity of the regions is expressed as a picture using ArcView GIS.

• The Korean Journal of Applied Statistics
• /
• v.23 no.5
• /
• pp.871-882
• /
• 2010

In animal tumorigenicity data, the occurrence time of tumor is not observed because the existence of a tumor is examined only at either time of natural death or time of sacrifice for the animal. A three-state model (Health-Tumor onset-Death) is widely used to model the incomplete data. In this paper, we employed a frailty effect into the three-state model to incorporate the dependency of death on tumor occurrence when the time of natural death works as an informative censoring against the tumor onset time. For the inference of parameters, then the EM algorithm is considered in order to deal with missing quantities of tumor onset time and random frailty. The proposed method is applied to the bladder tumor data taken from Lindsey and Ryan (1993, 1994) and a simulation study is performed to show the behavior of the proposed estimators.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.2
• /
• pp.57-64
• /
• 2014

Most previous studies of software fault prediction model which determines the fault-proneness of input modules have focused on supervised learning model using training data set. However, Unsupervised learning model is needed in case supervised learning model cannot be applied: either past training data set is not present or even though there exists data set, current project type is changed. Building an unsupervised learning model is extremely difficult that is why only a few studies exist. In this paper, we build unsupervised models using representative clustering algorithms, EM and DBSCAN, that have not been used in prior studies and compare these models with the previous model using K-means algorithm. The results of our study show that the EM model performs slightly better than the K-means model in terms of error rate and these two models significantly outperform the DBSCAN model.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.5
• /
• pp.583-597
• /
• 2001

Tracking the targets of interest has been one of the major research areas in radar surveillance system. We formulate the tracking problem as an incomplete data problem and apply the EM algorithm to obtain the MAP estimate. The resulting filter has a recursive structure analogous to the Kalman filter. The difference is that the measurement-update deals with multiple measurements and the parameter-update can estimate the system parameters. Through extensive experiments, it turns out that the proposed system is better than PDAF and NNF in tracking the targets. Also, the performance degrades gracefully as the disturbances become stronger.