http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
내장 문자와 사전 구조 지식을 이용한 HTMLtoVXML 변환 에이전트 개발
장영건,Jang, Young-Gun 한국정보처리학회 2003 정보처리학회논문지D Vol.10 No.2
본 연구는 음성 인터넷 서비스를 위하여 기존의 HTML 컨텐츠를 자동으로 VXML 컨텐츠로 변환하여 사용하는 방법에 관한 것이다. 문서 변환은 HTML 문서의 해석과 내용 분리에 앞서 컨텐츠의 선택이 필수적으로 요구되지만, 이미 알고 있고, 관심이 있는 내용의 집단적 선택에 대하여는 아직까지 좋은 해법이 없어 보인다. 본 논문에서는 비구조적 데이터로 구성된 웹 문서로부터 필요한 정보 묶음을 선택하기 위하여 웹 문서에 포함된 문자열, 구조에 대한 사용자의 사전 지식을 상호 반응적으로 사용하는 방법을 제안하며, 이를 구현하여 그 타당성을 입증하였다. 제안하는 방식은 기존의 구조적 방식에 의한 선택보다 사용자의 의도를 정확히 반영하며, 문서 표현의 구현 기법의 변화에 강건한 장점이 있다. 또한 컨텐츠 분리 측면에서는 XML또는 XHTML 문서변환을 중간 과정으로 사용하는 방식에 비하여 시간적으로 빠르며, 변환 과정의 부담이 작다. This paper presents a new agent which convert HTML contents to VXML contents automatically for voice services via web. In this paper, I propose an interactive hybrid sequential contents selection method to select desired contents fast and robustly from known web pages. It uses real time structural features as well as embedded text and/or priori structural knowledge such as link symbol position. To verify its effectiveness, a full agent system is implemented and tested. The method reflects user intention more accurately than conventional selections using structural features and is more robust to variations of HTML programming techniques. The agent is fast and has less computational burden than methods use XML or XHTML conversion as intermediate stage.
히스토그램과 퍼지 기법을 이용한 레이저 용접 결함 인식에 관한 연구
장영건,Jang, Young-Gun 한국전기전자학회 2001 전기전자학회논문지 Vol.5 No.2
본 논문은 용접이상을 검출하기 위한 특징벡터의 선택과 퍼지 기술을 사용한 용접이상 분류기의 설계 및 구현에 관한 것이다. 용접이상 특징 벡터로써 시간 영역에서 절대적분치, 영교차수를, 주파수 영역에서 파워 스펙트럼 계수를, 두 영역 모두를 고려하여 히스토그램을 비교하였다. 그래프 분석에 의하여 특징벡터로서 히스토그램을 선택하였고, 상대 히스토그램의 최대 빈도수와 대응 구간 값이 정상 용접과 용입불량을 구분하는 데 가장 유효하다는 것을 발견하였다. 이 특징 벡터를 사용하여 퍼지 용접이상 분류기를 구현하였고, 695개의 용접 데이터 프레임에 대하여 시험하여 정분류율이 92.96%을 보여, 그 유효성을 입증하였다. 실험실에서의 결과로써 실제적인 산업용 레이저 용접 검사기로써 상대적 히스토그램을 이용한 퍼지 용접이상 분류기가 효과적임을 알 수 있다. This paper is addressed to welding defect feature vector selection and implementation method of welding defect classifier using fuzzy techniques. We compare IAV, zero-crossing number as time domain analysis, power spectrum coefficient as frequency domain, histogram as both domain for welding defect feature selection. We choose histogram as feature vector by graph analysis and find out that maximum frequent occurrence number and section of corresponding signal scale in relative histogram show obvious difference between normal welding and voiding with penetration depth defect. We implement a fuzzy welding defect classifier using these feature vector, test it to verify its effectiveness for 695 welding data frame which consist of 4000 sampled data. As result of test, correct classification rate is 92.96%. Lab experimental results show a effectiveness of fuzzy welding defect classifier using relative histogram for practical Laser welding monitoring system in industry.
장영건 청주대학교 산업과학연구소 1999 産業科學硏究 Vol.17 No.1
An online plasma monitoring system which detect plasma signals and welding defects by analyzing the signals using photo detector is developed. In this paper, intelligent recognition method for welding defect is proposed using histogram as welding feature. To recognize welding defects effectively and reliably, one of the most impotant aspect is feature selection. Histogram of plasma signal intensity represent both of statistic and dynamic informations of the signals. Lab experimental results and welding experimental results show effectiveness of proposed method and correct welding defect recognition rate is 98.43%, 100% respectively using perceptron and fuzzy logic for 0.5 sec data, we import the welding defect recognition algorithm on a DSP board in the system.
장영건 청주대학교 산업과학연구소 2000 産業科學硏究 Vol.18 No.1
Automated guided vehicle systems (AGVS) are widely used for transporting material in manufacturing and warehousing applications. These systems offer many advantages over other forms of material transport. However, the design of these systems is complex due to the interrelated decisions that must be made and the large number of system design alternatives that are available. In particular, the design of the AGVS control system for one traction motor type can be quite challenging, and it can dramatically reduce the system cost and increase the performance. In this paper, an AGVS control system design is presented using traditional 2 traction motor type controller and new additional controller. It is convenient and simple for manufacturing and maintenance as well as maintaing advantages of one traction motor type controller.
레이져 용접의 결함을 인식하기 위한 특징벡터 추출에 관한 연구
장영건,조경환 청주대학교 산업과학연구소 2000 産業科學硏究 Vol.17 No.2
An online plasma monitoring system which detects plasma signals and recognizes welding defects by analyzing the signals using photo detector is developed. To recognize welding defects effectively and reliably, one of the most important things is feature extraction. In this paper, various time domain and frequency domain features for welding defect recognition are introduced, compared and analyzed considering target seperability, real time capability, memory reduction and noise effect compression. Simulation results and welding experimental results show maximum and corresponding relative signal intensity of histogram without nomalization is the best feature among zero crossing rate, IAV, PSD, LPC and other histogram features. Using the feature, correct recognition rate is over 92% with robustness, minimum memory usage and simple recognition method.
장영건 청주대학교 산업과학연구소 1996 産業科學硏究 Vol.14 No.-
A new neural network model, time delayed counterpropagation neural network(TDCPN) which has higher recognition rate but lower learning time, is proposed. The model has the time-delay arrangement of time delayed neural network(TDNN) and the learning method of Counter-propagation network(CPN). The performance of a proposed model in the recognition of numerical speechs uttered by specific speaker is tested. Experimental results show that the recognition rate of the model is better than CPN, the learning time is shorter than TDNN. But this model needs a lot of memory 40 times more than TDNN for high recognition.
장영건 청주대학교 산업과학연구소 2006 産業科學硏究 Vol.24 No.1
Dyslexia, learning disability can be medicated or overcome by using proper assertive devices. The dyslexics suffer from insufficient color contrast of electronic documents and mangle deficit of brain to access them. In this study, a clinical test for the algorithm of ARRT and WCAG 2.0 is performed to evaluate the effectiveness of the algorithm and threshold values. A new algorithm considering color temperature feeling is presented for dyslexics and S/W color filters are developed for those of Ireland syndrome. The results of the test show that each algorithm of WK has a different superiority and defect over each other, is not perfect and the suggested algorithm is superior to others with respect to linearity between color contrast and readability.