• 제어로봇시스템학회논문지
• /
• 제14권12호
• /
• pp.1278-1285
• /
• 2008

An optimization-based real-time joint angle extraction method of human elbow is proposed by processing the biomedical signal of surface EMG (electromyogram) measured at the center point of biceps brachii. The EMG signal is known as non-stationary (time-varying) signal, but we assume that it is quasi-stationary because a physical or physiological system has limitations in the rate at which it can change its characteristics. Based on the assumption, a pre-processing method to obtain pre-angle values from raw EMG signal is firstly suggested, and then an optimization method to minimize the error between the pre-angle and real joint angle is proposed in this paper. Finally, we suggest the experimental results showing the effectiveness of the proposed algorithm.

• 한국표면공학회:학술대회논문집
• /
• 한국표면공학회 2014년도 추계학술대회 논문집
• /
• pp.293-293
• /
• 2014

The applicability of graphene has been demonstrated in the electronic fields. But, high performance of graphene is limited by the contact resistance (Rc) at the metal-graphene interface. Recently, Rc was found to be improved by forming edge-contacted graphene via theoretical simulation. Based on the differences between the surface and edge contacts at the M-G interface, we demonstrate "edge-contacted" graphene through the use of a controlled plasma processing technique that generates the edge structure of the bond and significantly reduces the contact resistance. The contact resistance attained by using pre-plasma processing was of $270{\Omega}{\cdot}{\mu}m$. Mechanisms of pre-plasma process leading to low Rc was revealed by SEM and Raman spectroscopy. In the end, controlled pre-plasma processing enabled to fabricate CVD-graphene field effect transistors with an enhanced adhesion and improved carrier mobility.

• Journal of Information Processing Systems
• /
• 제18권2호
• /
• pp.282-291
• /
• 2022

This research applies a pre-trained bidirectional encoder representations from transformers (BERT) handwriting recognition model to predict foreign Korean-language learners' writing scores. A corpus of 586 answers to midterm and final exams written by foreign learners at the Intermediate 1 level was acquired and used for pre-training, resulting in consistent performance, even with small datasets. The test data were pre-processed and fine-tuned, and the results were calculated in the form of a score prediction. The difference between the prediction and actual score was then calculated. An accuracy of 95.8% was demonstrated, indicating that the prediction results were strong overall; hence, the tool is suitable for the automatic scoring of Korean written test answers, including grammatical errors, written by foreigners. These results are particularly meaningful in that the data included written language text produced by foreign learners, not native speakers.

• 한국컴퓨터정보학회논문지
• /
• 제27권2호
• /
• pp.15-23
• /
• 2022

특허 분야에서 자연어처리(Natural Language Processing) 태스크는 특허문헌의 언어적 특이성으로 문제 해결의 난이도가 높은 과제임에 따라 한국 특허문헌에 최적화된 언어모델의 연구가 시급한 실정이다. 최근 자연어처리 분야에서는 특정 도메인에 특화되게 사전 학습(Pre-trained)한 언어모델을 구축하여 관련 분야의 다양한 태스크에서 성능을 향상시키려는 시도가 지속적으로 이루어지고 있다. 그 중, ELECTRA는 Google이 BERT 이후에 RTD(Replaced Token Detection)라는 새로운 방식을 제안하며 학습 효율성을 높인 사전학습 언어모델이다. 본 연구에서는 대량의 한국 특허문헌 데이터를 사전 학습한 KorPatELECTRA를 제안한다. 또한, 특허 문헌의 특성에 맞게 학습 코퍼스를 정제하고 특허 사용자 사전 및 전용 토크나이저를 적용하여 최적화된 사전 학습을 진행하였다. KorPatELECTRA의 성능 확인을 위해 실제 특허데이터를 활용한 NER(Named Entity Recognition), MRC(Machine Reading Comprehension), 특허문서 분류 태스크를 실험하였고 비교 대상인 범용 모델에 비해 3가지 태스크 모두에서 가장 우수한 성능을 확인하였다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2010년도 춘계학술대회 논문집
• /
• pp.1335-1336
• /
• 2010

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(4)
• /
• pp.70-73
• /
• 2000

The error diffusion is good for reproducing continuous image to binary image. However the reproduction of edge characteristics is weak in power spectrum analysis of display error. It is suggested for us an edge-enhanced error-diffusion method that is included pre-processing algorithm for edge characteristic enhancement. Pre-processing algorithm is organized horizontal and vertical directional 2nd order differential values and weighting function of pre-filter. The improved Error diffusion using pre-filter, presents a good results visually which edge characteristics is enhanced. The performance of the proposed algorithm is compared with that of the conventional edge-enhanced error diffusion by measuring the RAPSD of display error, the egde correlation and the local average accordance.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2002년도 학술대회지
• /
• pp.67-70
• /
• 2002

In the present study a pre/post-processor program has been developed to be used with a general CFD code. This program is capable of performing the basic functions of the pre/post-processing, which include mesh generation and post processing plots. Also through perspective projection, this program can be used to check the quality of generated mesh by moving around inside the mesh. The smoke visualization can be also performed with the present program to visualize the smoke behavior in the case of fire simulation. The examples of the program execution are given in paper.

• 한국자동차공학회논문집
• /
• 제8권3호
• /
• pp.190-197
• /
• 2000

A graphic vehicle modeling pre-processing program and a visualization post-processing program have been developed for AutoDyn7, which is a special program for vehicle dynamics. The Rapid-App for GUI(Graphic User Interface) builder and the Open Inventor for 3D graphic library have been employed to develop these programs in Silicon Graphics workstation. A Graphic User Interface program integrates vehicle modeling pre-processor, AutoDyn7 analysis processor, and visualization post-processor. In vehicle modeling pre-processor, vehicle hard point data for a suspension model are automatically converted into multibody vehicle system data. An interactive graphics capabilities provides suspension modeling aides to verify user input data interactively. In visualization post-processor, vehicle virtual test simulation results are animated with virtual testing environments.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 하계학술대회 논문집 B
• /
• pp.1289-1291
• /
• 1996

In this paper, we propose a new adaptive pre-processing technique for the linearization of a weakly nonlinear system which can be modeled by a Volterra series up to third order. To compensate the nonlinear effects of a given system, an update algorithm for the linear filter coefficients of the proposed adaptive pre-processor is introduced, and to compensate the linear distortion of the given system, the linear inverse filter is also utilized. For the performance test of the proposed adaptive pre-processor, computer simulation results obtained by analyzing an ANRSS loudspeaker model are provided.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제9권4호
• /
• pp.145-152
• /
• 2020

기계독해는(Machine reading comprehension) 사용자 질의와 관련된 문서를 기계가 이해한 후 정답을 추론하는 인공지능 자연어처리 태스크를 말하며, 이러한 기계독해는 챗봇과 같은 자동상담 서비스에 활용될 수 있다. 최근 자연어처리 분야에서 가장 높은 성능을 보이고 있는 BERT 언어모델은 대용량의 데이터를 pre-training 한 후에 각 자연어처리 태스크에 대해 fine-tuning하여 학습된 모델로 추론함으로써 문제를 해결하는 방식이다. 본 논문에서는 BERT기반 특허상담 기계독해 태스크를 위해 특허상담 데이터 셋을 구축하고 그 구축 방법을 소개하며, patent 코퍼스를 pre-training한 Patent-BERT 모델과 특허상담 모델학습에 적합한 언어처리 알고리즘을 추가함으로써 특허상담 기계독해 태스크의 성능을 향상시킬 수 있는 방안을 제안한다. 본 논문에서 제안한 방법을 사용하여 특허상담 질의에 대한 정답 결정에서 성능이 향상됨을 보였다.