• 대한전자공학회논문지
• /
• 제27권5호
• /
• pp.757-766
• /
• 1990

In this paper a new knowledge based vision system using orientation information on each surface of the 3-dimensional object is discussed. The measurement of the orientation information is performed by photometric stereo method. And then the obtained orientations are segmented using Gaussian curvature and mean curvature. A hierarchical knowledge base which is based on the characteristics, shape, area and length of the surface is built up, and then the knowledge based system infers by the condition interprete system (CIS). As the results, an easier and more accurate 3-D object recognition system is implemented, because it uses the characteristics and shapes as units of the surface in the recognition process.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2000년도 추계학술대회 논문집
• /
• pp.199-202
• /
• 2000

In this paper, we developed an expert system that helped a machine-tool designer to select a moderate servomotor of the feed drive unit of a machine tool. With the specification of the drive unit, it searches for the set of servo-motors that have enough torque and power to satisfy the driving condition which the designer defines from the database. me database also contains knowledge and rules, which describe the design process and calculate design parameters of a feed drive. The knowledge-based design support module shows every steps of inference and reason for the solution that it provides.

• 한국주거학회:학술대회논문집
• /
• 한국주거학회 1995년도 학술발표대회논문집 하
• /
• pp.85-93
• /
• 1995

Recently, the development of Artificial Intelligence(AI) and Expert System has caused some interest in the possibility of developing an intelligent CAD system. However, these development and possibility are in an extremenly early stage for Architectural design. In this study, the design process of Residence being recognized as a Constraints-satisfied model, a part of these constraints used in the Architectural design of Residence are being systematized and sorted by the design process. Those regulations and planning items to be considered in the basic planning stage are being systematized as a knowledge base system. The possibility of this knowledge-based system as an effective design tool is studied and an integrated form of Architectural design system is proposed.

• 한국정보통신학회논문지
• /
• 제10권9호
• /
• pp.1691-1699
• /
• 2006

본 논문에서는 모든 분야의 지식 에 대해 온톨로지(ontology)화 하기보다는 회의 분야로 특화하여 회의 지을 온톨로지화 할 수 있는 지식기반회의 관리시스템 아키텍처를 제시한다. 또한, 시뮬레이션 및 분석을 통해 객관적 평가 기준을 마련한 다음 화상 회의, 음성 인식, 행동 인식 등의 각종 인식 시스템을 통해 변환되고 가공된 정보를 체계화하고 지식 화하여 분석 할 수 있는 시스템으로의 발전 가능성에 대해서 제시한다.

• 한국콘크리트학회:학술대회논문집
• /
• 한국콘크리트학회 1993년도 가을 학술발표회 논문집
• /
• pp.71-78
• /
• 1993

The R/C structures is very popular because of its low construction cost and its semi-permanent life. But in Korea, unfortunately there are few exprets in this field, and the repair methods and selection of material for R/C structures are determined by their subjective personal opinions. Therefore systematic study of this field is ulgently required. For attainning this purpose, in this study the related documents and knowledge or experience from human experts were collected. Based on the collected information cracks are classified into 25 patterns and repair related-knowledge base, which will be formulated and encoded into the domain knowledge, is built. And then an expert system, that can suggest the repair methods in the same way the human experts would, is developed. The results using the developed expert system are compared to the real field practices and they are satisfactory.

• Asia pacific journal of information systems
• /
• 제2권1호
• /
• pp.35-56
• /
• 1992

This paper proposes a synergism of neural networks (NN) and knowledge-based decision support system (KBDSS) to effectively design an intelligent strategic planning system. Since conventional KBDSS becomes inoperative partially or totally when problem deviates slightly from the expected problem-domain, a new DSS concept is needed for designing an effective strategic planning system, where strategic planning environment is usually turbulent and consistently changing. In line with this idea, this paper developes a NN-based DSS, named ConDSS, incorporating the generalization property of NN into its knowledge base. The proposed ConDSS was extensively operated in an experimentally designed environment with three models: BCG matrix, Growth/Gain matrix, and GE matrix. The results proved very promising when encountered with unforeseen situations in comparisons with conventional KBDSS.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 1999년도 춘계학술대회논문집
• /
• pp.213-217
• /
• 1999

An expert system is developed for rational and efficient design of multi-component injection molding which is a fairly new manufacturing technique to produce plastic parts by injecting two or more materials sequentially using multiple injection units in a single machine into a single rotary mold. The knowledge base used in the present design system is primarily composed of two parts ; knowledge from domain expert and knowledge from CAE analysis. The present expert system has hour main modules ; general design guidelines for injection molding specific guidelines for multi-component injection molding redesign guidelines from the result of the CAE analysis and finally troubleshooting for multi-component injection molding. To show the validity of the present design methodology two shop floor design problems were tested ; design and fabrication of timing belt cover and power window's assist knob by using multi-component injection molding.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅳ
• /
• pp.1871-1874
• /
• 2003

Knowledge-based numeric open caption recognition is proposed that can recognize numeric captions generated by character generator (CG) and automatically superimpose a modified caption using the recognized text only when a valid numeric caption appears in the aimed specific region of a live sportscast scene produced by other broadcasting stations. in the proposed method, mesh features are extracted from an enhanced binary image as feature vectors, then a valuable information is recovered from a numeric image by perceiving the character using a multiplayer perceptron (MLP) network. The result is verified using knowledge-based hie set designed for a more stable and reliable output and then the modified information is displayed on a screen by CG. MLB Eye Caption based on the proposed algorithm has already been used for regular Major League Base-ball (MLB) programs broadcast five over a Korean nationwide TV network and has produced a favorable response from Korean viewer.

• 한국정밀공학회지
• /
• 제16권10호
• /
• pp.217-224
• /
• 1999

Trouble shooting of modern machine tools equipped with sophisticated electronic as well as mechanical parts is so difficult that it is usually depends upon the experience and accumulated knowledge of the diagnosing persons. On the other hand, tool users are scattered in wide area, which makes it expensive for a machine tool maker to run a vast service network. An unmanned diagnosis system to which users can have access at all times could be an efficient alternative. For this purpose, a rule-based expert system for diagnosing machine tools is developed. This paper describes the structure of diagnostic knowledge, the rule firing mechanism, the diagnosis flow, and user query process. An example shows the feasibility of problem solving on site without help of a service expert from machine tool maker.

• 전자공학회논문지B
• /
• 제31B권7호
• /
• pp.119-128
• /
• 1994

In this paper, we propose a method for facial feature extraction and recognition algorithm using neural networks. First we extract a face part from the background image based on the knowledge that it is located in the center of an input image and that the background is homogeneous. Then using vertical and horizontal projections. We extract features from the separated face image using knowledge base of human faces. In the recognition step we use the back propagation algorithm of the neural networks and in the learning step to reduce the computation time we vary learning and momentum rates. Our technique recognizes 6 women and 14 men correctly.