• 대한의용생체공학회:의공학회지
• /
• 제11권2호
• /
• pp.243-248
• /
• 1990

EMG (eleltromyographic) signals are generated by contracting muscle and detected in and out side of muscle in the form of random signals. In the measurement of muscle fatigue, the mean frequency of EMG signals using spectrum analysis is an important parameter in diagonosis of muscle disease and in sports medicine fields. In this study, the degree of spectral transfer to lower frequency caused by accumulation of Latic acid inside the muscle is estimated. The new spectral analysis method using 2"d order hAaximum Entropy Method was applied to estimate the mean frequency and we confirmed that this new method yields fast and reliable estimation.tion.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권2호
• /
• pp.1118-1133
• /
• 2017

Computer vision-based human activity recognition (HAR) has become very famous these days due to its applications in various fields such as smart home healthcare for elderly people. A video-based activity recognition system basically has many goals such as to react based on people's behavior that allows the systems to proactively assist them with their tasks. A novel approach is proposed in this work for depth video based human activity recognition using joint-based motion features of depth body shapes and Deep Belief Network (DBN). From depth video, different body parts of human activities are segmented first by means of a trained random forest. The motion features representing the magnitude and direction of each joint in next frame are extracted. Finally, the features are applied for training a DBN to be used for recognition later. The proposed HAR approach showed superior performance over conventional approaches on private and public datasets, indicating a prominent approach for practical applications in smartly controlled environments.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2011년도 한국컴퓨터종합학술대회논문집 Vol.38 No.1(C)
• /
• pp.258-261
• /
• 2011

Living in the age of information people face problems related to information overload. Information is easy to produce, store and distribute through various communication channels, one of which is emails. With the appearance of the mobile devices, such as smart phones and tabs, people can have access to email inbox at any moment of time from everywhere. In this paper we present information extraction system with a specific goal of extracting meeting location from the announcement of seminar or conference. We apply a machine learning method (conditional random fields, CRF), train the system using annotated corpus of seminar and conference announcements and validate results by applying various extracted correction rules and patterns. Furthermore, we normalize extracted location, and reference using geo-coding databases, OpenStreetMap and Wikipedia resources to determine real geographical coordinates.

• 디지털산업정보학회논문지
• /
• 제16권3호
• /
• pp.49-58
• /
• 2020

Image interpolation, a technology that converts low resolution images into high resolution images, has been widely used in various image processing fields such as CCTV, web-cam, and medical imaging. This technique is based on the fact that the statistical distributions of the white Gaussian noise and the difference between the interpolated image and the original image is similar to each other. The proposed algorithm is composed of three steps. In first, the interpolated image is derived by random image interpolation. In second, we derive weighting functions that are used to apply non-local mean filtering. In the final step, the prediction error is corrected by performing non-local mean filtering by applying the selected weighting function. It can be considered as a post-processing algorithm to further reduce the prediction error after applying an arbitrary image interpolation algorithm. Simulation results show that the proposed method yields reasonable performance.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2008년도 제20회 한글 및 한국어 정보처리 학술대회
• /
• pp.23-28
• /
• 2008

한국어 처리에서 구문분석기에 대한 요구는 많은 반면 성능의 한계와 강건함의 부족으로 인해 채택되지 못하는 것이 현실이다. 본 연구는 구문분석을 레이블링 문제로 전환하여 성능, 속도, 강건함을 모두 실현한 시스템에 대해서 설명한다. 우리는 다단계 구 단위화(Cascaded Chunking)를 통해 한국어 구문분석을 시도한다. 각 단계에서는 어절별 품사 태그와 어절 구문표지를 자질로 사용하고 Conditional Random Fields(CRFs)를 이용하여 최적의 결과를 얻는다. 98,412문장 세종 구문 코퍼스로 학습하고 1,430문장(평균 14.59어절)으로 실험한 결과 87.30%의 구문 정확도를 보였다. 이 결과는 기존에 제안되었던 구문분석기와 대등하거나 우수한 성능이며 기존 구문분석기가 처리하지 못하는 장문도 처리 가능하다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1999년도 가을 학술발표회 논문집
• /
• pp.351-358
• /
• 1999

The structural responses of underground structures are examined in probability by using the elasto-plastic stochastic finite element method in which the spatial distributions of material properties are assumed to be stochastic fields. In addition, the adaptive importance sampling method using the response surface technique is used to improve simulation efficiency. The method is found to provide appropriate information although the nonlinear Limit State involves a large number of basic random variables and the failure probability is small. The probability of plastic local failures around an excavated area is effectively evaluated and the reliability for the limit displacement of the ground is investigated. It is demonstrated that the adaptive importance sampling method can be very efficiently used to evaluate the reliability of a large scale stochastic finite element model, such as the underground structures located in the multi-layered ground.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2009년도 제21회 한글 및 한국어 정보처리 학술대회
• /
• pp.87-91
• /
• 2009

전문용어의 수가 급증하면서 전문용어를 자동으로 인식하는 연구가 활발히 진행되고 있다. 전문용어를 인식하기 위해서 전문용어의 범위를 정한 뒤 그 전문용어의 분야를 선택해야 한다. 본 논문에서는 생물/의학 사전정보와 CRF(Conditional Random Fields) 기계학습 기법을 사용하여 연구를 진행한다. 기계학습을 위한 자질로 품사, 접사, 대소문자, 숫자, 특수문자, 단서어휘 등을 사용한다. 특히 단서어휘와 사전정보를 중요한 요소로 생각하여, 3가지 방법으로 나누어 실험한다. 총 분야의 개수는 7개이며, 각 분야별로 정확률, 재현율, F-measure를 측정한다. 경계인식은 83.92%의 정확률, 96.42%의 재현율, 89.73의 F-measure가 결과로 나타났고, 분야분류는 79.29%의 정확률, 91.06%의 재현율, 84.77%의 F-measure가 결과로 나타났다.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2009년도 제21회 한글 및 한국어 정보처리 학술대회
• /
• pp.186-190
• /
• 2009

품사 부착은 각종 자연어처리의 기본적인 요소이며, 크게 규칙 기반 방법과, 통계 기반 방법으로 나눌 수 있다. 대부분은 통계 기반의 기계학습을 이용하고 있으며, 대개 95% 이상의 성능을 보여주고 있다. 그러나 미등록어에 대해서는 성능이 그다지 높지 않다. 이 논문에서는 단어의 접사 정보를 이용해서 미등록어에 대한 품사 부착의 성능을 높이는 방법을 제안한다. 제안된 시스템은 CRF(Conditional Random Fields)를 이용하며, 그 자질의 일부로 접사 정보를 이용한다. 그 결과 미등록어에 대해서 약 40%의 성능이 개선되었다. 앞으로 미등록어에 적합한 자질을 연구하고 개발할 필요가 있을 것으로 생각된다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1678-1680
• /
• 2002

Stereo matching technique is used in many practical fields like satellite image analysis and computer vision. In this paper, we suggest a method to extract a target object image from a complicated background. For example, human face image can be extracted from random background. This method can be applied to computer vision such as security system, dressing simulation by use of extracted human face, 3D modeling, and security system. Many researches about stereo matching have been performed. Conventional approaches can be categorized into area-based and feature-based method. In this paper, we start from area-based method and apply area tracking using scanning window. Coarse depth information is used for area merging process using area searching data. Finally, we produce a target object image.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1993년도 하계학술대회 논문집 B
• /
• pp.932-934
• /
• 1993

Optimum pole shape is designed to increase the levitation and propulsion force of magnetic levitation systems. Evolution Strategy is introduced as optimization method. Evolution Strategy is random based non-deterministic method, developed by combining Genetic Algorithm with Simulated Annealing. Trasnsrapid-06, which was developed in Germany, is referenced model to be analyze. Design variables are nodes which determine fields pole shape of a linear synchronous motor, and the model analyzed by F.E.M.