• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국제학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.573-574
• /
• 1992

We developed a recognition and tracking system of moving objects. The system consists of one CCD video camera, two DC motors in horizontal and vertical axles with encoders, pluse width modulation(PWM) driving unit, 16 bit NEC 9801 microcomputer, and their interfaces. The recognition and tracking system is able to recognize shape and size of a moving object and is able to track the object within a certain range of errors. This paper presents the brief introduction of the recognition and tracking system developed in our laboratory.

• 한국산업정보학회논문지
• /
• 제7권3호
• /
• pp.74-80
• /
• 2002

한국어 연속 음성 인식결과의 성능향상을 위해서 자연어 처리 기술을 이용한 후처리 기법이 사용된다. 그러나 자연어 처리 기법이 대부분 띄어쓰기가 있는 정형화된 입력 문장에 대한 분석을 수행하여 왔기 때문에 형태소 분석기를 직접 음성인식 결과의 향상에 사용하는 데에는 어려운 점이 많다. 본 논문에서는 띄어쓰기를 고려하지 않는 기능어 기반의 최장일치 형태소 해석 방법인 구문 형태소 단위의 분석을 이용한 음정인식 결과의 향상 모델을 제안한다. 제안된 모델을 통해 연속음성 인식 결과에서 자주 발생하는 용언과 보조 용언 및 의존 명사 사이의 음운들 사이의 구조적 정보를 활용함으로써 음성 인식 결과의 성능을 향상시키는 방법에 대해 기술한다.

• 디지털콘텐츠학회 논문지
• /
• 제10권1호
• /
• pp.121-127
• /
• 2009

자동차 ECU(Electronic Control Unit)는 날이 갈수록 더욱 복잡해지고 많은 기능을 요구하고 있다. 대표적으로 power windows switch, LCM(Light Control Module), mirror control system, seat memory등 운전자 편의 시스템이 개발되어 양산 중에 있다. 또한 현재 업계에서 많은 연구개발이 진행되고 있는 운전자 편의를 위한 DIS(Driver Information System)도 있다. 하지만 이러한 시스템을 운전 중 조작하게 되면 많은 위험이 따른다. 따라서 본 논문에서는 이러한 자동차 편의장치를 음성으로 조작 가능한 음성인식 시스템을 구현하였으며 자동차 잡음환경에서 인식률 향상을 위한 전처리 필터를 적용하여 양호한 인식결과 얻었다.

• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.730-744
• /
• 2023

The topic of this paper is the recognition of human activities using egocentric vision, particularly captured by body-worn cameras, which could be helpful for video surveillance, automatic search and video indexing. This being the case, it could also be helpful in assistance to elderly and frail persons for revolutionizing and improving their lives. The process throws up the task of human activities recognition remaining problematic, because of the important variations, where it is realized through the use of an external device, similar to a robot, as a personal assistant. The inferred information is used both online to assist the person, and offline to support the personal assistant. With our proposed method being robust against the various factors of variability problem in action executions, the major purpose of this paper is to perform an efficient and simple recognition method from egocentric camera data only using convolutional neural network and deep learning. In terms of accuracy improvement, simulation results outperform the current state of the art by a significant margin of 61% when using egocentric camera data only, more than 44% when using egocentric camera and several stationary cameras data and more than 12% when using both inertial measurement unit (IMU) and egocentric camera data.

• 대한안전경영과학회지
• /
• 제13권3호
• /
• pp.55-61
• /
• 2011

Unit cost for asbestos removal work which have been made out at government contract award for recent three years has been reviewed and shown with table. And the cost have been compared with ones made out by two agencies, American asbestos removal specialists, government ministry for making out comparison tables. First, legal and practical work status survey have been made to determine reasonability of introducing separate contract-awarding system, as a part of ensuring reasonable unit work cost. And then, two different status have been compared and there in introduction possibility of separate contract-awarding system, it is found out. In interior removal work case, it is thought that 50% by owner, 60% unit work cost by the expert of which removal specialist think as reasonable unit cost. the results for exterior and spray work method are shown with almost same context. Impact factors for determining unit work cost such as project mount, project kind, contract-awarding method, subcontract method have been compared with parties. Removal specialist and expert group have almost same ideas for project amount which has been hightestly weighed but owner's recognition has partially lower than specialists idea. There are almost no difference of recognition between three(3) parties for project kind. Idea comparison of reasonable unit work cost for asbestos removal work between three concerned parties has been analyzed and found out that expert group do unit work cost, 65% of which removal specialists estimate, in interior tex removal. And there is almost no difference between them in exterior roof slate removal. But there is considerable difference between them that owner estimates 50% unit cost of specialist one while expert group estimate 50% of specialist idea of unit removal cost.

• 융합정보논문지
• /
• 제10권8호
• /
• pp.35-39
• /
• 2020

DNN은 기존의 음성 인식 시스템에 비해 에러가 적으나 병렬 훈련이 어렵고, 계산의 양이 많으며, 많은 양의 데이터 확보를 필요로 한다. 본 논문에서는 이러한 문제를 효율적으로 해결하기 위해 GMM에서 모델 파라메터를 가지고 음소별 GMM 파라메터를 추정하여 음소 단위를 생성한다. 그리고 이를 효율적으로 적용하기 위해 특정 어휘에 대한 클러스터링을 통해 성능을 향상시키기 위한 방법을 제안한다. 이를 위해 3가지 종류의 단어 음성 데이터베이스를 이용하여 DB를 가지고 어휘 모델을 구축하였고, 잡음 처리는 워너필터를 사용한 특징을 추출하여 음성 인식실험에 사용하였다. 본 논문에서 제안한 방법을 사용한 결과 음성 인식률에서 97.9%의 인식률을 나타내었다. 본 연구에서 개선된 오버피팅의 문제점을 향상시킬 수 있는 추가적인 연구를 필요로 한다.

• 한국주거학회논문집
• /
• 제5권2호
• /
• pp.51-63
• /
• 1994

The purpose of this study are to examine the types and sources of information used for housing choices and to figure out the related factors. Data are collected through self-administered questionnaires designed for this study, and the sample consists of 396 households in Pusan and Ulsan. The purposes are accomplished through descriptive statistics and multiple regression analyses. Based on the results of analysis, housing information are divided into four specific types : economic. technical, housing unit, and neighborhood information. It is found that housing unit information including housing quality and economic information such as housing prices are identified as the most important ones for current and future housing choices. The most useful sources of housing information utilized for current housing are friends, relatives and neighbors. In addition, model house, real estate office, newspaper and adverizement are the another useful sources for housing information. Among them, the model house is the most helpful one for variety of housing information. Young households and those with a head whose occupation is professional/managerial tend to have higher recognition of the importance of housing information than do the other groups. The households currently living in small apartment and with a young eldest child are likely to have higher recognition of the importance of econimic information. Tenure type, the occupation of the household head, the age of the eldest child, and length of residence are the significant explanatory variables of the recognition of the importance of housing unit information.

• Bulletin of the Korean Chemical Society
• /
• 제31권6호
• /
• pp.1551-1554
• /
• 2010

Two diastereomeric chiral stationary phases (CSPs) were applied to the liquid chromatographic resolution of various racemic ${\alpha}$-amino methyl esters, ${\alpha}$-amino N,N-diethylamides and ${\alpha}$-amino N-propylamides. The CSP incorporating (R)-3,3'-diphenyl-1,1'-binaphtyl and (R,R)-tartaric acid unit as chiral barriers did not show any chiral recognition. In contrast, the CSP incorporating (R)-3,3'-diphenyl-1,1'-binaphtyl and (S,S)-tartaric acid unit as chiral barriers was found to show excellent chiral recognition especially for the two enantiomers of ${\alpha}$-amino N-propylamides. Some of ${\alpha}$-amino methyl esters and ${\alpha}$-amino N,N-diethylamides were also resolved on the CSP incorporating (R)-3,3'-diphenyl-1,1'-binaphtyl and (S,S)-tartaric acid unit. From these results it was concluded that the two chiral units composing the diastereomeric CSPs can show "matched" or "mismatched" effect on the chiral recognition according to their absolute stereochemistry.

• 한국정밀공학회지
• /
• 제10권4호
• /
• pp.142-151
• /
• 1993

An intelligent mulitiple monitoring system to monitor tool/machining states synthetically was proposed and developed. It consists of 2 fundamental subsystems : the multiple sensor detection unit and the intellignet integrated diagnosis unit. Three signals, that is, spindle motor current, Z-axis motor current, and machining sound were adopted to detect tool/machining states more reliably. Based on the multiple sensor information, the diagnosis unit judges either tool breakage or degree of tool wear state using fuzzy reasoning. Tool breakage is diagnosed by the level of spindle/z-axis motor current. Tool wear is diagnosed by both the result of fuzzy pattern recognition for motor currents and the result of pattern matching for machining sound. Fuzzy c-means algorithm was used for fuzzy pattern recognition. Experiments carried out for drill operation in the machining center have shown that the developed system monitors abnormal drill/states drilling very reliably.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2008년도 춘계종합학술대회 A
• /
• pp.784-788
• /
• 2008

In this research, we propose a remote speech input unit, a new method of user-friendly speech input in speech recognition system. We focused the user friendliness on hands-free and microphone independence in speech recognition applications. Our module adopts two algorithms, the automatic speech detection and speech enhancement based on the microphone array-based beamforming method. In the performance evaluation of speech detection, within-200msec accuracy with respect to the manually detected positions is about 97percent under the noise environments of 25dB of the SNR. The microphone array-based speech enhancement using the delay-and-sum beamforming algorithm shows about 6dB of maximum SNR gain over a single microphone and more than 12% of error reduction rate in speech recognition.