• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2006.10a
• /
• pp.143-144
• /
• 2006

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2006.10a
• /
• pp.477-478
• /
• 2006

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.3
• /
• pp.472-478
• /
• 2012

The inspection machine in PCB (printed circuit board) assembly line checks assembly errors by inspecting the images inside of the component inspection region. The component inspection region consists of region of component package and region of soldering. It is necessary to extract the regions automatically for auto-teaching system of the inspection machine. We propose an image segmentation method to extract the component inspection regions automatically from images of PCB. The acquired image is transformed to HSI color model, and then segmented by several regions by clustering method. We develop a modified K-means algorithm to increase the accuracy of extraction. The heuristics generating the initial clusters and merging the final clusters are newly proposed. The vertical and horizontal projection is also developed to distinguish the region of component package and region of soldering. The experimental results are presented to verify the usefulness of the proposed method.

• Proceedings of the KSME Conference
• /
• 2007.05a
• /
• pp.1559-1563
• /
• 2007

This paper presents a virtual machine design simulation program. Kinematics of various mechanisms can be modeled with 3 dimensional geometry and actuators. CAD data for any machine component can be easily imported in STL format. Machine components are assembled with kinematic joints simply by drag and drop function in virtual graphic simulator. Interference and collision of any component with other components can be identified during the motion simulation. Graphic user interface program is developed using Microsoft Direct X code. A precision micro stage system is demonstrated with the proposed virtual machine design simulator.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.3
• /
• pp.57-64
• /
• 2016

This paper deals with the PCB assembly time minimization problem that the PAP (pick-and-placement) machine pickup the K-weighted group of N-components, loading, and place into the PCB placement location. This problem considers the rotational turret velocity according to component weight group and moving velocity of distance in two component placement locations in PCB. This paper suggest heavy-weight component group first pick-and-place strategy that the feeder sequence fit to the placement location Hamiltonean cycle sequence. This algorithm applies the quadratic assignment problem (QAP) that considers feeder sequence and location sequence, and the linear assignment problem (LAP) that considers only feeder sequence. The proposed algorithm shorten the assembly time than iATMA for QAP, and same result as iATMA that shorten the assembly time than ATMA.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.7
• /
• pp.1288-1295
• /
• 2002

The column-type sensing element in building and mechanical construction parts was designed as three forces and three moments sensor by attaching strain gages approximately. Compared to conventional multi-component sensor, the designed sensor has high stiffness and low cost. The radius of the column was designed analytically and compared with finite element analysis. The interference errors between components were minimized by using addition and subtraction procedure of signals. The fabricated sensor was tested by using a deadweight force standard machine and a six-component force calibration machine. The calibration results showed that the 6-component forces and moments sensor had interference error less than 7.3 % between $F_x$ and $M_x$ components, and 5.0 % in case of other components.

• Proceedings of the KSME Conference
• /
• 2001.06a
• /
• pp.837-841
• /
• 2001

The column-type sensing element in building and mechanical construction parts was designed as three forces and three moments sensor by attaching strain gages approximately. Compared to conventional multi-component sensor, the designed sensor can solve the problem about low stiffness and high cost. The radius of the column was designed analytically and compared with finite element analysis. The coupling errors between components were minimized by using addition and subtraction procedure of signals. The fabricated sensor was tested by using a deadweight force standard machine and a six-component force calibration machine in Korea Research Institute of Standards and Science(KRISS). The calibration showed that the multi-component force/moment sensor had coupling error less than 19.8 % between $F_x$ and $M_y$ components, and 9.0 % in case of other components.

• ETRI Journal
• /
• v.37 no.3
• /
• pp.541-550
• /
• 2015

In this paper, we propose a classification-based approach for hybridizing statistical machine translation and rulebased machine translation. Both the training dataset used in the learning of our proposed classifier and our feature extraction method affect the hybridization quality. To create one such training dataset, a previous approach used auto-evaluation metrics to determine from a set of component machine translation (MT) systems which gave the more accurate translation (by a comparative method). Once this had been determined, the most accurate translation was then labelled in such a way so as to indicate the MT system from which it came. In this previous approach, when the metric evaluation scores were low, there existed a high level of uncertainty as to which of the component MT systems was actually producing the better translation. To relax such uncertainty or error in classification, we propose an alternative approach to such labeling; that is, a cut-off method. In our experiments, using the aforementioned cut-off method in our proposed classifier, we managed to achieve a translation accuracy of 81.5% - a 5.0% improvement over existing methods.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.453-456
• /
• 2022

The role of the power grid is to ensure stable power supply. It is necessary to take various measures to prepare for unstable situations without notice. After identifying the relationship between features through exploratory data analysis using weather data, a machine learning based energy production prediction model is modeled. In this study, the prediction reliability was increased by extracting the features that affect energy production prediction using principal component analysis and then applying it to the machine learning model. By using the proposed model to predict the production energy for a specific period and compare it with the actual production value at that time, the performance of the energy production prediction applying the principal component analysis was confirmed.

• Journal of the Korea Safety Management & Science
• /
• v.5 no.4
• /
• pp.187-194
• /
• 2003

The purpose of a man-machine chart is improved utilization of a man or a machine. Improved utilization can mean less idle time, rebalanced idle time, or less idle time of an excessive component. In this study we improved the productivity of the assembly line in the electronics company using man-machine chart. The results are applicable for improving the effectiveness of line balancing problems in the electronics industry.