• Journal of Advanced Navigation Technology
• /
• v.27 no.6
• /
• pp.888-896
• /
• 2023

When an aircraft is landing, it is expected that the aircraft will follow a specified approach procedure and then land at the airport. However, depending on the airport situation, neighbouring aircraft or the instructions of the air traffic controller, there can be a deviation from the specified approach. Detecting aircraft approach patterns is necessary for traffic flow and flight safety, and this paper suggests clustering techniques to identify aircraft patterns in the approach segment. The Gaussian Mixture Model (GMM), one of the machine learning techniques, is used to cluster the trajectories of aircraft, and ADS-B data from aircraft landing at the Gimhae airport in 2019 are used. The aircraft trajectories are clustered on the plane, and a total of 86 approach trajectory patterns are extracted using the centroid value of each cluster. Considering the correlation between the approach procedure pattern and overshoots, the distribution of overshoots is calculated.

• Journal of Intelligence and Information Systems
• /
• v.15 no.3
• /
• pp.53-65
• /
• 2009

The part-of-speech determination is necessary for resolving the part-of-speech ambiguity in English-Korean machine translation. The part-of-speech ambiguity causes high parsing complexity and makes the accurate translation difficult. In order to solve the problem, the resolution of the part-of-speech ambiguity must be performed after the lexical analysis and before the parsing. This paper proposes the CatAmRes model, which resolves the part-of-speech ambiguity, and compares the performance with that of other part-of-speech tagging methods. CatAmRes model determines the part-of-speech using the probability distribution from Bayesian network training and the statistical information, which are based on the Penn Treebank corpus. The proposed CatAmRes model consists of Calculator and POSDeterminer. Calculator calculates the degree of appropriateness of the partof-speech, and POSDeterminer determines the part-of-speech of the word based on the calculated values. In the experiment, we measure the performance using sentences from WSJ, Brown, IBM corpus.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.2
• /
• pp.288-295
• /
• 2011

The paper proposes a localization method of an underwater robot using Monte Carlo Localization(MCL) approach. Localization is one of the fundamental basics for autonomous navigation of an underwater robot. The proposed method resolves the problem of accumulation of position error which is fatal to dead reckoning method. It deals with uncertainty of the robot motion and uncertainty of sensor data in probabilistic approach. Especially, it can model the nonlinear motion transition and non Gaussian probabilistic sensor characteristics. In the paper, motion model is described using Euler angles to utilize the MCL algorithm for position estimation of an underwater robot. Motion model and sensor model are implemented and the performance of the proposed method is verified through simulation.

• Information and Communications Magazine
• /
• v.24 no.4
• /
• pp.27-35
• /
• 2007

음성 통신을 할 때 배경 잡음이 존재하게 되면 일반적으로 음질이 저하된다. 이것은 잡음 자체가 듣기 싫다거나 음성을 더 작게 들리게 만들기 때문이기도 하고 음성 코덱이 잡음이 섞이지 않은 깨끗한 음성에 최적화되어 있어서 잡음이 섞인 음성에 대한 코딩 효율이 떨어지기 때문이기도 하다. 이 논문에서는 잡음에 의한 음성 통신의 품질 저하를 막기 위한 방법으로서 음성 향상(speech enhancement) 기술과 음성 강화(speech reinforcement) 기술에 대해 소개한다. 음성 향상 기술이란 전송부의 마이크에서 녹음된 잡음과 음성이 섞인 입력 음성으로부터 깨끗한 음성을 추정하는 기술을 말한다. 음성 향상 기술은 상당히 오랜 기간 동안 연구되어 온 기술이며, 최근에는 각 파라미터의 분포에 의존하는 방법보다 확률 모델에 기반한 방법이 각광을 받고 있으며 인간의 청각 특성을 고려한 음성 향상 방법도 제안되고 있다. 음성 강화 기술이란 수신단에서 주변 잡음에 따라 전송되어 온 음성을 주파수별로 증폭하여 더 잘 들리도록 만드는 기술이다. 음성 향상이 내 주위의 잡음이 상대방에게 들리는 음성에 미치는 영향 혹은 상대방 주변의 잡음이 나에게 들리는 소리에 미치는 영향을 줄여주는 기술이라면 음성 강화는 내 주위의 잡음이 나에게 들리는 음성에 미치는 영향을 상쇄해 주는 기술이다. 이 경우 주변 잡음은 어떤 전자 시스템도 거치지 않고 귀로 직접 들어오기 때문에 잡음 자체를 줄여 주는 것은 힘들고 전송되어 온 음성을 적절히 증폭 혹은 변형함으로써 귀에 들리는 음질 또는 명료성을 개선하게 된다. 이 논문에서는 통계 모델을 기반으로 한 음성 향상 기법과 인간의 청각 특성을 고려한 음성 향상 기법, 그리고 음성 강화 기법에 대해 설명한다.을 시도한 결과 안정적이고 반복 가능한 급성 심부전 모델을 얻을 수 있었다. bench scale실험결과와 같이 AOC는 배수관망에서의 박테리아 증식과 크게 상관관계를 갖고 있는 것으로 밝혀졌다.)', 'have a headache (2.10±0.79)', 'poor memory (2.09±0.83)', 'no appetite (1.99±0.85)', As for the correlation between iron parameter and clinical symptoms related to anemia, the hematocrit rate was negatively correlated with 'get a cold easily', 'pale face', 'feeling blue', 'difficult digestion' (p<0.05). The level of iron was negatively correlated with 'tired out easily', 'get a cold easily' (p<0.05) and TS (%) were negatively correlated with 'tired out easily (p<0.05)', 'get a cold easily (p<0.01). Our study resulted that the prevalence of a iron deficiency of a middle school girl is very high, therefore the guidelines for iron supplementation and nutritional education to improve their iron status should be provided.한 질소제거를 N-balance로부터

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.4
• /
• pp.512-517
• /
• 2011

In this paper, we propose an Active Shape Model (ASM) and Lucas-Kanade (LK) optical flow-based feature extraction and analysis method for analyzing the emotional features from facial images. Considering the facial emotion feature regions are described by Facial Action Coding System, we construct the feature-related shape models based on the combination of landmarks and extract the LK optical flow vectors at each landmarks based on the centre pixels of motion vector window. The facial emotion features are modelled by the combination of the optical flow vectors and the emotional states of facial image can be estimated by the probabilistic estimation technique, such as Bayesian classifier. Also, we extract the optimal emotional features that are considered the high correlation between feature points and emotional states by using common spatial pattern (CSP) analysis in order to improvise the operational efficiency and accuracy of emotional feature extraction process.

• Journal of the Korea Computer Graphics Society
• /
• v.26 no.5
• /
• pp.15-24
• /
• 2020

We present a supervised learning method that estimates the simulation parameters required to simulate the fabric from the static drape shape of a given fabric sample. The static drape shape was inspired by Cusick's drape, which is used in the apparel industry to classify fabrics according to their mechanical properties. The input vector of the training model consists of the feature vector extracted from the static drape and the density value of a fabric specimen. The output vector consists of six simulation parameters that have a significant influence on deriving the corresponding drape result. To generate a plausible and unbiased training data set, we first collect simulation parameters for 400 knit fabrics and generate a Gaussian Mixed Model (GMM) generation model from them. Next, a large number of simulation parameters are randomly sampled from the GMM model, and cloth simulation is performed for each sampled simulation parameter to create a virtual static drape. The generated training data is fitted with a log-linear regression model. To evaluate our method, we check the accuracy of the training results with a test data set and compare the visual similarity of the simulated drapes.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.27 no.2
• /
• pp.103-111
• /
• 2014

In this study, the probabilistic internal pressure fragility analysis was performed by using the non-linear finite element analysis method. The target structure is one of the containment buildings of typical domestic pressurized water reactors(PWRs). The 3-dimensional finite element model of the containment building was developed with considering the large equipment hatches. To consider uncertainties in the material properties and structural capacities, we performed the sensitivity analysis of the ultimate pressure capacity with respect to the variation of four important uncertain parameters. The results of the sensitivity analysis were used to the selection of the probabilistic variables and the determination of their probabilistic parameters. To reflect the present condition of the tendon pre-stressing force, the data of the pre-stressing force acquired from the in-service inspections of tendon forces were used for the determination of the median value. Two failure modes(leak, rupture) were considered and their limit states were defined to assess the internal pressure fragility of target containment building. The internal pressure fragilities for each failure mode were evaluated in terms of median internal pressure capacity, high confidence low probability of failure(HCLPF) capacity, and fragility curves with respect to the confidence levels. The HCLPF capacity was 115.9 psig for leak failure mode, and 125.0 psig for rupture failure mode.

• Korean Journal of Construction Engineering and Management
• /
• v.24 no.6
• /
• pp.45-52
• /
• 2023

Recently, as climate change caused by greenhouse gases is intensifying, the international community has committed to reduce greenhouse gas emissions. The purpose of this study is to present the methodology and major considerations for investment judgment. Two actual cases of overseas projects were selected as study subjects. As an analysis method, the major risk factors were defined as a probability distribution, and the NPV was stochastically estimated using the Monte Carlo simulation method. In addition, assuming a policy change, the range of NPV change was analyzed. As a result, the average NPV of project A was lowered by 19%, and the probability of showing a negative NPV was 12.2%. The average value of project B was lowered by 12.5%. Considering the policy change, project A can obtain economic benefits only when it obtains 72.9% or more of the total amount of carbon credits generated, and project B is economically feasible when it acquires 49.5% or more. As a result, the average value of project A is lower than the net present value under basic assumptions, so caution is needed in investment decisions depending on changes in major risk factors. Additionally, considering policy changes, the carbon credit distribution ratio should be differentially applied depending on the project size, and this was presented as a specific figure.

• Journal of Food Hygiene and Safety
• /
• v.28 no.1
• /
• pp.36-40
• /
• 2013

This study was performed to determine the ranking foods and related foodborne pathogens for the impact of climate change, e.g., increasing temperature, on microbiological food safety. To do this, we developed an impact-ranking model comprising an Excel spreadsheet by using Risk Ranger. Because of a lack of data, input values in this model were determined on the basis of an expert's opinion. These values also were converted to normal distribution, and the developed model was simulated using @RISK. In conclusion, the 5 superior ranking foods and related foodborne pathogens for climate change impact were as follows: ready-to-eat foods (RTE) (Staphylococcus aureus, Salmonella spp., and Escherichia coli O157:H7); bread and rice cakes (S. aureus and Bacillus cereus); meat and egg products (Salmonella spp., E. coli O157:H7, and S. aureus); tofu (bean curds) and jellies (B. cereus, E. coli O157:H7, and S. aureus); and fish products (S. aureus, Vibrio spp., and E. coli O157:H7).

• Journal of KIISE
• /
• v.42 no.4
• /
• pp.451-458
• /
• 2015

Previous multimodal learning methods focus on problem-solving aspects, such as image and video search and tagging, rather than on knowledge acquisition via content modeling. In this paper, we propose the Multimodal Concept Hierarchy (MuCH), which is a content modeling method that uses a cartoon video dataset and a character-based subtitle generation method from the learned model. The MuCH model has a multimodal hypernetwork layer, in which the patterns of the words and image patches are represented, and a concept layer, in which each concept variable is represented by a probability distribution of the words and the image patches. The model can learn the characteristics of the characters as concepts from the video subtitles and scene images by using a Bayesian learning method and can also generate character-based subtitles from the learned model if text queries are provided. As an experiment, the MuCH model learned concepts from 'Pororo' cartoon videos with a total of 268 minutes in length and generated character-based subtitles. Finally, we compare the results with those of other multimodal learning models. The Experimental results indicate that given the same text query, our model generates more accurate and more character-specific subtitles than other models.