• Journal of Korea Society of Industrial Information Systems
• /
• v.3 no.1
• /
• pp.141-151
• /
• 1998

The importance of fax imagerestoration is constantly increasing with the widespreasd use of facsimile machines in almost every sector of modern society. Recently, Handley and Doutherty proposed a morphological method for restoration of fax images, Their method removed effectively the only salient noise in the fax images. However, it could not remove the white and pepper noise that can appear in fax images since they treated fax images as deterministic sequences rather than random fields. Furthermore, this approach suffers from computational burden since it does not use recursive restoration technique. To cope with those difficulties, in this paper, we propose a new restoration scheme for restoring fax images using Kalman fitering which provides and efficient recursive processor. The proposed restoration method is based on the wide-sense Markov random fields (WSM).In order to verify the performance of the proposed restoration method, several experiments with the CCITT Group 3 fax machine were conducted with the generated document .Experimental results revealed that our proposed restoration method was shown to be superior to Handley et.al's method for restoring fax images.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.9
• /
• pp.4405-4418
• /
• 2016

Automatic image completion techniques have difficulty processing images in which the target region has multiple planes or is non-facade. Here, we propose a new image completion method that uses belief propagation based on planar priorities. We first calculate planar information, which includes planar projection parameters, plane segments, and repetitive regularity extractions within the plane. Next, we convert this planar information into planar guide knowledge using the prior probabilities of patch transforms and offsets. Using the energy of the discrete Markov Random Field (MRF), we then define an objective function for image completion that uses the planar guide knowledge. Finally, in order to effectively optimize the MRF, we propose a new optimization scheme, termed Planar Priority-belief propagation that includes message-scheduling-based planar priority and dynamic label cropping. The results of experiment show that our approach exhibits advanced performance compared with existing approaches.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.172-175
• /
• 2018

환자의 병력을 서술하는 임상문서에서 임상 개체명들은 그들 사이에 개체명이 아닌 단어들이 위치하기 때문에 거리상으로 서로 떨어져 있고, 임상 개체명인식에 많이 사용되는 조건부무작위장(conditional random fields; CRF) 모델은 Markov 속성을 따르기 때문에 서로 떨어져 있는 개체명 라벨 간의 전이 정보는 모델의 계산에서 무시된다. 본 논문에서는 라벨링 모델에 서로 떨어진 개체명 간 전이 관계를 표현하기 위하여 CRF 모델의 구조를 변경하는 방법론을 소개한다. 제안된 CRF 모델 디자인에서는 모델의 계산효율성을 빠르게 유지하기 위하여 Markov 속성을 유지하는 1차 모델 구조를 유지한다. 모델은 선행하는 개체명의 라벨 정보를 후행하는 개체명 엔터티에게 전달하기 위하여 선행 개체명의 라벨을 뒤 따르는 비개체명 라벨에 전이시키고 이를 통해 후행하는 개체명은 선행하는 개체명의 라벨 정보를 알 수 있게 된다. 라벨의 고차 전이 정보를 전달함에도 모델의 구조는 1차 전이 구조를 유지함으로 n차 구조의 모델보다 빠른 계산 속도를 유지할 수 있게 된다. 모델의 성능 평가를 위하여 서울대학교병원 류머티즘내과에서 퇴원한 환자들의 퇴원요약지에 병력과 관련된 엔터티가 태깅된 평가 데이터와 i2b2 2012/VA 임상자연어처리 shared task의 임상 개체명 추출 데이터를 사용하였고 기본 CRF 모델들(1차, 2차)과 비교하였다. 피처 조합에 따라 모델들을 평가한 결과 제안한 모델이 거의 모든 경우에서 기본 모델들에 비하여 F1-score의 성능을 향상시킴을 관찰할 수 있었다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.5 s.311
• /
• pp.37-44
• /
• 2006

In this paper, a new version of the ICM method(MICM, modified ICM) in which the contextual information is modelled by Markov random fields (MRF) is introduced. To extract the feature, a new local MRF model with a fitting block neighbourhood is proposed. This model selects contextual information not only from the relative intensity levels but also from the geometrically directional position of neighbouring cliques. Feature extraction depends on each block's contribution to the local variance. They discriminates it into several regions, for example context and background. Boundaries between these regions are also distinctive. The proposed algerian performs segmentation using directional block fitting procedure which confines merging to spatially adjacent elements and generates a partition such that pixels in unified cluster have a homogeneous intensity level. From experiment with ink rubbed copy images(Takbon, 拓本), this method is determined to be quite effective for feature identification. In particular, the new algorithm preserves the details of the images well without over- and under-smoothing problem occurring in general iterated conditional modes (ICM). And also, it may be noted that this method is applicable to the handwriting recognition.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.8
• /
• pp.2555-2564
• /
• 2000

In this paper, a new unsupervised image segmentation algorithm is proposed. To model the contextual information presented in images, the characteristics of the Markov random fields (MRF) are utilized. Textured images are modeled as realizations of the stationary Gaussian MRF on a two-dimensional square lattice using the conditional autoregressive (CAR) equations with a second-order noncausal neighborhood. To detect boundaries, hypothesis tests over two masked areas are performed. Under the hypothesis, masked areas are assumed to belong to the same class of textures and CAR equation parameters are estimated in a minimum-mean-square-error (MMSE) sense. If the hypothesis is rejected, a measure of dissimilarity between two areas is accumulated on the rejected area. This approach produces potential edge maps. Using these maps, boundary detection can be performed, which resulting no micro edges. The performance of the proposed algorithm is evaluated by some experiments using real images as weB as synthetic ones. The experiments demonstrate that the proposed algorithm can produce satisfactorY segmentation without any a priori information.

• The Korean Journal of Applied Statistics
• /
• v.22 no.2
• /
• pp.425-433
• /
• 2009

Spatial models suitable for describing the evolving random fields in climate and environmental systems have been developed by many researchers. In general, rainfall in South Korea is highly variable in intensity and amount across space. This study characterizes the monthly and regional variation of rainfall fields using the spatial modeling. The main objective of this research is spatial prediction with the Bayesian hierarchical modeling (kriging) in order to further our understanding of water resources over space. We use the Bayesian approach in order to estimate the parameters and produce more reliable prediction. The Bayesian kriging also provides a promising solution for analyzing and predicting rainfall data.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.2
• /
• pp.118-126
• /
• 2008

Statistical image reconstruction methods have played an important role in emission computed tomography (ECT) since they accurately model the statistical noise associated with gamma-ray projection data. Although the use of statistical methods in clinical practice in early days was of a difficult problem due to high per-iteration costs and large numbers of iterations, with the development of fast algorithms and dramatically improved speed of computers, it is now inevitably becoming more practical. Some statistical methods are indeed commonly available from nuclear medicine equipment suppliers. In this paper, we first describe a mathematical background for statistical reconstruction methods, which includes assumptions underlying the Poisson statistical model, maximum likelihood and maximum a posteriori approaches, and prior models in the context of a Bayesian framework. We then review a recent progress in developing fast iterative algorithms.

• The Korean Journal of Applied Statistics
• /
• v.31 no.2
• /
• pp.287-301
• /
• 2018

It is common to encounter count data with excess zeros in various research fields such as the social sciences, natural sciences, medical science or engineering. Such count data have been explained mainly by zero-inflated Poisson model and extended models. Zero-inflated count data are also often correlated or clustered, in which random effects should be taken into account in the model. Frequentist approaches have been commonly used to fit such data. However, a Bayesian approach has advantages of prior information, avoidance of asymptotic approximations and practical estimation of the functions of parameters. We consider a Bayesian zero-inflated Poisson regression model with random effects for correlated zero-inflated count data. We conducted simulation studies to check the performance of the proposed model. We also applied the proposed model to smoking behavior data from the Regional Health Survey (2015) of the Korea Centers for disease control and prevention.

• Communications for Statistical Applications and Methods
• /
• v.23 no.6
• /
• pp.497-515
• /
• 2016

Modern processes often monitor more than one quality characteristic that are referred to as multivariate statistical process control (MSPC) procedures. The MSPC is the most rapidly developing sector of statistical process control and increases interest in the simultaneous inspection of several related quality characteristics. Most multivariate detection procedures based on a multi-normality assumptions are independent, but there are many processes that assume non-normality and correlation. Many multivariate control charts have a lack of related joint distribution. Copulas are tool to construct multivariate modelling and formalizing the dependence structure between random variables and applied in several fields. From copula literature review, there are a few copula to apply in MSPC that have multivariate control charts, and represent a successful tool to identify an out-of-control process. This paper presents various types of copulas modelling for the multivariate control chart. The performance measures of the control chart are the average run length (ARL) and the average number of observations to signal (ANOS). Furthermore, a Monte Carlo simulation is shown when the observations were from an exponential distribution.

• ETRI Journal
• /
• v.45 no.5
• /
• pp.822-835
• /
• 2023

Recent semantic segmentation frameworks usually combine low-level and high-level context information to achieve improved performance. In addition, postlevel context information is also considered. In this study, we present a Context ReFinement Network (CRFNet) and its training method to improve the semantic predictions of segmentation models of the encoder-decoder structure. Our study is based on postprocessing, which directly considers the relationship between spatially neighboring pixels of a label map, such as Markov and conditional random fields. CRFNet comprises two modules: a refiner and a combiner that, respectively, refine the context information from the output features of the conventional semantic segmentation network model and combine the refined features with the intermediate features from the decoding process of the segmentation model to produce the final output. To train CRFNet to refine the semantic predictions more accurately, we proposed a sequential training scheme. Using various backbone networks (ENet, ERFNet, and HyperSeg), we extensively evaluated our model on three large-scale, real-world datasets to demonstrate the effectiveness of our approach.