• KIPS Transactions on Software and Data Engineering
• /
• 제4권7호
• /
• pp.297-300
• /
• 2015

In medical ultrasonography, transit time and contrast enhancement patterns are considered as important parameters to analyze liver diseases. In many recent researches, time-intensity curves(TIC) have been used for calculating the transit time of the contrast agents. However, the intensity curve may include the variations which are caused by the micro-bubble effect of contrast agents. In this paper, we propose a complementary approach to diagnostic parameter extraction which utilizes a density information as well as the intensity data. The proposed technique improves the accuracy in extraction of the transit time and velocity of contrast agents for detection and characterization of focal liver lesions. Through the experiments using a set of clinical data, we show that the proposed methods can improve the reliability of the parametric image data.

• Journal of KIISE
• /
• 제42권7호
• /
• pp.910-918
• /
• 2015

This paper presents a parameter visualization technique to overcome the limitation of the naked eye in contrast-enhanced ultrasonography. A method is also proposed to compensate for the distortion and noise in ultrasound image sequences. Meaningful parameters for diagnosing liver disease can be extracted from the dynamic patterns of the contrast enhancement in ultrasound images. The visualization technique can provide more accurate information by generating a parametric image from the dynamic data. Respiratory motions and noise from micro-bubble in ultrasound data may cause a degradation of the reliability of the diagnostic parameters. A multi-stage algorithm for respiratory motion tracking and an image enhancement technique based on the Markov Random Field are proposed. The usefulness of the proposed methods is empirically discussed through experiments by using a set of clinical data.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• 제18권2호
• /
• pp.179-194
• /
• 2020

A Bayesian approach was introduced to improve the belief of prior distributions of input parameters for the probabilistic safety assessment of radioactive waste repository. A GoldSim-based module was developed using the Markov chain Monte Carlo algorithm and implemented through GSTSPA (GoldSim Total System Performance Assessment), a GoldSim template for generic/site-specific safety assessment of the radioactive repository system. In this study, sequential Bayesian updating of prior distributions was comprehensively explained and used as a basis to conduct a reliable safety assessment of the repository. The prior distribution to three sequential posterior distributions for several selected parameters associated with nuclide transport in the fractured rock medium was updated with assumed likelihood functions. The process was demonstrated through a probabilistic safety assessment of the conceptual repository for illustrative purposes. Through this study, it was shown that insufficient observed data could enhance the belief of prior distributions for input parameter values commonly available, which are usually uncertain. This is particularly applicable for nuclide behavior in and around the repository system, which typically exhibited a long time span and wide modeling domain.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• 제14권4호
• /
• pp.107-112
• /
• 2014

This Study is Transfer function optimization plan of volume rendering of multi user environment. Each volume data, for appropriate transfer function, they should be adjusted parameter many times. To prevent this, we propose transfer function optimization plan using crowd sourcing. In multi user environment, we use weight value for reliability level for each user. Because transfer function parameter used previous users is provided next users, they can be used effectively optimized transfer function and can reduce attempts.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• 제36권11호
• /
• pp.1072-1078
• /
• 2008

An adaptive control algorithm for spacecraft rendezvous and docking in a Keplerian orbit is presented. The equations of relative motion of two spacecrafts expressed in a local-vertical-local-horizontal rectangular frame are converted to a general Hamiltonian form, then an adaptive control method developed for the uncertain Hamiltonian system is applied to the rendezvous and docking problem. A smooth projection algorithm is applied to keep the parameter estimates inside a singularity-free region, and a numerical example shows that the developed controller successfully deals with the unknown mass of the chaser spacecraft.

• Proceedings of the KIEE Conference
• /
• 대한전기학회 2003년도 하계학술대회 논문집 D
• /
• pp.2567-2569
• /
• 2003

본 논문은 비선형 시스템의 퍼지모델을 위해 정보 Granules 기반 퍼지추론 시스템 모델의 최적화를 제시한다. 퍼지모델은 주로 경험적 방법에 의해 추출되기 때문에 보다 구체적이고 체계적인 방법에 의한 동정 및 최적화 될 필요성이 요구된다. 제안된 규칙베이스 퍼지모델은 HCM 클러스터링 방법, 컴플렉스 알고리즘 및 퍼지추론 방법을 이용하여 시스템 구조와 파라미터 동정을 수행한다. 두 가지 형태의 퍼지모델 추론 방법은 간략추론, 선형추론에 의해 시행된다. 본 논문에서는 퍼지모델의 입력변수와 퍼지 입력 공간 분할 및 입출력 데이타의 중심값을 구해서 후반부 다항식함수에 의한 정보 Granules 기반 구조 동정과 파라미터 동정을 통해 비선형 시스템을 표현한다. 전반부 파라미터의 동정에는 HCM 클러스터링 방법과 컴플렉스 알고리즘을 사용하고, 후반부는 표준 HCM 클러스터링과 표준 최소자승법을 사용하여 동정한다. 그리고 학습 및 테스트 데이타의 성능견과의 상호균형을 얻기 위한 하중값을 가진 성능지수를 제시함으로써 근사화와 예측성능의 향상을 꾀한다. 제안된 비선형 모델의 성능평가를 통해 그 우수성을 보인다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 한국정보처리학회 2023년도 추계학술발표대회
• /
• pp.514-517
• /
• 2023

본 연구는 인 컨택스트 러닝 (In-Context Learning)을 오디오-언어 작업에 적용하기 위한 멀티모달 (Multi-Modal) 딥러닝 모델을 다룬다. 해당 모델을 통해 학습 단계에서 오디오와 텍스트의 소통 가능한 형태의 표현 (Representation)을 학습하고 여러가지 오디오-텍스트 작업을 수행할 수 있는 멀티모달 딥러닝 모델을 개발하는 것이 본 연구의 목적이다. 모델은 오디오 인코더와 언어 인코더가 연결된 구조를 가지고 있으며, 언어 모델은 6.7B, 30B 의 파라미터 수를 가진 자동회귀 (Autoregressive) 대형 언어 모델 (Large Language Model)을 사용한다 오디오 인코더는 자기지도학습 (Self-Supervised Learning)을 기반으로 사전학습 된 오디오 특징 추출 모델이다. 언어모델이 상대적으로 대용량이기 언어모델의 파라미터를 고정하고 오디오 인코더의 파라미터만 업데이트하는 프로즌 (Frozen) 방법으로 학습한다. 학습을 위한 과제는 음성인식 (Automatic Speech Recognition)과 요약 (Abstractive Summarization) 이다. 학습을 마친 후 질의응답 (Question Answering) 작업으로 테스트를 진행했다. 그 결과, 정답 문장을 생성하기 위해서는 추가적인 학습이 필요한 것으로 보였으나, 음성인식으로 사전학습 한 모델의 경우 정답과 유사한 키워드를 사용하는 문법적으로 올바른 문장을 생성함을 확인했다.

• Proceedings of the KIEE Conference
• /
• 대한전기학회 1997년도 하계학술대회 논문집 F
• /
• pp.2062-2065
• /
• 1997

컨버터나 인버터의 운전에 있어서 비간섭 전류제어는 그 시스템의 성능에 직접적으로 영향을 주는 중요한 요인이다. 하지만 일반적인 비간섭 제어방식은 모터나 입력 인덕터에 파리미터 오차가 있으면 잘 동작하지 않는다 이 논문에서는 PI 형태의 새로운 비간섭제어 방식을 제안하였다. 이 방식의 가장 큰 장점은 파라미터 오차에 대해 강인성을 가지는 것이다. 시뮬레이션을 통하여 제안된 제어기의 성능을 살펴보았다.

• Journal of the Korean Institute of Intelligent Systems
• /
• 제11권6호
• /
• pp.491-499
• /
• 2001

In this paper, the scheme of an efficient fuzzy rule generation and fuzzy system construction using GA(genetic algorithm) and FCM(fuzzy c-means) clustering algorithm is proposed for TSK(Takagi-Sugeno-Kang) type fuzzy system. In the structure identification, input data is transformed by PCA(Principal Component Analysis) to reduce the correlation among input data components. And then, a set fuzzy rules are generated for a given criterion by FCM clustering algorithm . In the parameter identification premise parameters are optimally searched by GA. On the other hand, the consequent parameters are estimated by RLSE(Recursive Least Square Estimate) to reduce the search space. From this one can systematically obtain the valid number of fuzzy rules which shows satisfying performance for the given problem. Finally, we applied the proposed method to the Box-Jenkins data and rice taste data modeling problems and obtained a better performance than previous works.

• Journal of the Korean Institute of Intelligent Systems
• /
• 제14권2호
• /
• pp.130-135
• /
• 2004

This paper presents a new design methods of the short-term load forecasting system (STLFS) using the data mining. The structure of the proposed STLFS is divided into two parts: the Takagi-Sugeno (T-S) fuzzy model-based classifier and predictor The proposed classifier is composed of the Gaussian fuzzy sets in the premise part and the linearized Bayesian classifier in the consequent part. The related parameters of the classifier are easily obtained from the statistic information of the training set. The proposed predictor takes form of the convex combination of the linear time series predictors for each inputs. The problem of estimating the consequent parameters is formulated by the convex optimization problem, which is to minimize the norm distance between the real load and the output of the linear time series estimator. The problem of estimating the premise parameters is to find the parameter value minimizing the error between the real load and the overall output. Finally, to show the feasibility of the proposed method, this paper provides the short-term load forecasting example.