• Journal of Institute of Control, Robotics and Systems
• /
• v.19 no.4
• /
• pp.364-371
• /
• 2013

Emotion Recognition is one of the important part to develop in human-human and human computer interaction. In this paper, we have focused on the performance of multi-class SVM (Support Vector Machine) with Gaussian RFB (Radial Basis function) kernel, which has been used to solve the problem of emotion recognition from physiological signals and to improve the accuracy of emotion recognition. The experimental paradigm for data acquisition, visual-stimuli of IAPS (International Affective Picture System) are used to induce emotional states, such as fear, disgust, joy, and neutral for each subject. The raw signals of acquisited data are splitted in the trial from each session to pre-process the data. The mean value and standard deviation are employed to extract the data for feature extraction and preparing in the next step of classification. The experimental results are proving that the proposed approach of multi-class SVM with Gaussian RBF kernel with OVO (One-Versus-One) method provided the successful performance, accuracies of classification, which has been performed over these four emotions.

• The Journal of Bigdata
• /
• v.2 no.2
• /
• pp.95-104
• /
• 2017

In 2014, Seoul Metropolitan Government launched a response service aimed at responding promptly to civil complaints. The complaints received are categorized based on their content and sent to the department in charge. If this part can be automated, the time and labor costs will be reduced. In this study, we collected 17,700 cases of complaints for 7 years from June 1, 2010 to May 31, 2017. We compared the XGBoost with RandomForest and confirmed the suitability of Korean text classification. As a result, the accuracy of XGBoost compared to RandomForest is generally high. The accuracy of RandomForest was unstable after upsampling and downsampling using the same sample, while XGBoost showed stable overall accuracy.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2000.11a
• /
• pp.643-646
• /
• 2000

In this study, we constructed a preview-sensing visual sensor system for weld seam tracking in real time in GMA welding. A sensor part consists of a CCD camera, a band-pass filter, a diode laser system with a cylindrical lens, and a vision board for inter frame process. We used a commercialized robot system which includes a GMA welding machine. To extract the weld seam we used a inter frame process in vision board from that we could remove the noise due to the spatters and fume in the image. Since the image was very reasonable by using the inter frame process, we could use the simplest way to extract the weld seam from the image, such as first differential and central difference method. Also we used a moving average method to the successive position data of weld seam for reducing the data fluctuation. In experiment the developed robot system with visual sensor could be able to track a most popular weld seam, such as a fillet-joint, a V-groove, and a lap-joint of which weld seam include planar and height directional variation.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2000.11a
• /
• pp.295-298
• /
• 2000

In this research, a novel sampling strategy for a CMM to inspect freeform surfaces is proposed. Unlike primitive surfaces, it is not easy to determine the number of sampling points and their locations for inspecting freeform surfaces. Since a CMM operates with slower speed in measurement than optical measuring devices, it is important to optimize the number and the locations of sampling points in the inspection process. When a complete inspection of a surface is required, it becomes more critical. Among various factors to cause shape errors of a final product, curvature characteristic is essential due to its effect such as stair-step errors in rapid prototyping and interpolation errors in NC tool paths generation. Shape errors are defined in terms of the average and standard deviation of differences between an original model and a produced part. Proposed algorithms determine the locations of sampling points by analyzing curvature distribution of a given surface. Based on the curvature distribution, a surface area is divided into several sub-areas. In each sub-area, sampling points are located as further as possible. The optimal number of sub-areas. In each sub-area, sampling points are located as further as possible. The optimal number os sub-areas is determined by estimating the average of curvatures. Finally, the proposed method is applied to several surfaces that have shape errors for verification.

• The KSFM Journal of Fluid Machinery
• /
• v.19 no.5
• /
• pp.20-27
• /
• 2016

The Francis-99 is a workshop initiated by the Norwegian University of Science and Technology (NTNU), Norway, and Lulea University of Technology (LTU), Sweden, in order to further validate the capabilities of the CFD technologies. The goal of the first workshop is to determine the state of the art of numerical predictions for steady operating conditions. When performing the CFD analysis, some geometry details are often neglected. In case of Francis Turbine, labyrinth seals are usually not include in the simulation domain, this may lead to inaccurate prediction of turbine efficiency. In this study, the CFD analysis for Francis-99 Workshop model has been performed for full domain of machine including top and bottom labyrinth seals. The efficiency value and distribution of velocity and pressure have been investigated and compared to the experimental data obtained from NTNU. By comparing the results, it was found that: With the top and bottom labyrinth seals in the domain, the CFD result was significantly improved in prediction of efficiency at all the operating point, especially at part load.

• Journal of information and communication convergence engineering
• /
• v.8 no.5
• /
• pp.591-594
• /
• 2010

Control devices perform various works for us in many areas. The device is being utilized for precision movement of certain object. In as much as control devices are activated by means of motors, motor control is important.[1][2] Generally, servo motors capable of precision control are more frequently used than DC motors. Use of 3 motors allows 3- way movement. Medical controllers for surgical operation require high precision. [3][4][5][6] AX-12, a servo motor can realize various types of movement. AX-12 can be easily manufactured in the form of a robotic arm and has features that MCU and its peripheral circuits are simple. For precision movement, 3 motors can be controlled by use of a single joystick and 2 buttons, with movement angles being adjusted by having preset values in the program changed.[7][8] By virtue of this study, we have realized small precision robotic arm system utilizing single joystick and 2 buttons. This system can control the robotic arm in the direction desired by the user. The system has been designed such that a joystick controls 2 motors with the remaining motor being controlled by a button. Single MCU is tasked with both control and movement.[9] We have shown precision robotic arm system in the Figure contained in the conclusion part and made reference to results of analysis in there. It has also been demonstrated that the system can be utilized in the industry.[8]

• The Journal of Korean Institute for Practical Engineering Education
• /
• v.4 no.1
• /
• pp.57-64
• /
• 2012

For design of a electrical Machine it is used conventional design methode to determine size with considering of magnetical circuit and windings. After this process, Flux density and Torque are calculated with FEM program. But most electrical machines are complicated in their configuration, therefore it is not easy to understand the design process and analyzing methode. We need to develop a educational material. In this paper using Linear force motor, which is relatively not complicated in their configuration and easy to understand the relationship between electro magnetics and force, will be explained the design process and calculating process with Finite element methode. And with FEM program will be calculated and illustrated flux density in each part and Force of the Linear force motor easy to understand.

• Journal of Korean Home Economics Education Association
• /
• v.7 no.2
• /
• pp.113-123
• /
• 1995

The purpose of this study is to investigate teachers’ recognition in Food/Nutrition, Textile/Clothing part in Home Economics Text Book of Middle School and to provide the basic data for the improvement of its curriculum. 147 Home Economics teachers in Taegu city and Kyungsangbukdo area responded to the questionnaire. The results are summarized as follows: 1. Most of Home Economics teachers have graduated Dept. of Home Economics Education and have ever taken teacher training. And even those who ever taken teacher training are not satisfied with training curriculum contents. Therefore, the result of this study shows that teacher training curriculum contents should be improved so as to be helpful for the actual teaching and learning. 2. In terms of the suitability of contents of food & nutrition and contents of textiles & clothing to the student’s learning development levels, the degree of suitability is in the order of nutrition & health, nutrition in adolescence, food selection, kinds and functions of nutrients in food & nutrition curriculum, and in the order of suitable clothing, mixture rate of fabrics, purchase of clothing, clothing in adolescence, clothing selection. The contents of making processed foods and usage of sewing machine of the existing text book have turned out not to be appropriate. 3. Most teachers suggest that dietary guideline for health, misconception about food & nutrition selection of ready-made suit suitable clothing for situation & character as well as the contents of the existing text book should be included in the new text book.

• Proceedings of the KSME Conference
• /
• 2003.04a
• /
• pp.1014-1020
• /
• 2003

We have developed the ultrasonic polishing system to get super finishing that consist of machine part that can rotate and travel the main shaft with power 1.5kW, ultrasonic generator with frequency 20kHz. By using this system we were investigated the characteristics of ultrasonic polishing and deduced the major facters which affect the surface roughness by the experimental plans for three different materials such as ceramic, glass, and wafer, and so could be obtained following results. We could be obtained the excellent surface for hard-to-difficult cutting materials. The rotating speed could be found to be major factor influencing the surface roughness. In the case of ceramic and wafer, we were able to obtain good surface roughness when the feed rate and ultrasonic output is higher. In the case of glass, the surface roughness becames worse when ultrasonic output is higher because of increasing of load affacting on the particles in slurry.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.16 no.3
• /
• pp.1-8
• /
• 2017

Tool path generation is a part of process planning in 3D printing. This is done before actual printing by a computer rather than an AM machine. The mesh geometry of the 3D model is sliced layer-by-layer along the Z-axis and tool paths are generated from the sliced layers. Each 2-dimensional layer can have two types of printing paths: (i) shell and (ii) infill. Shell paths are made of offset loops. During shell generation, twists can be produced in offset loops which will cause twisted tool paths. As a twisted tool path cannot be printed, it is necessary to remove these twists during process planning. In this research, An algorithm is presented to remove twists from the offset loops. To do so the path segments are traversed to identify twisted points. Outer offset loops are represented in the counter-clockwise segment order and clockwise rotation for the inner offset loop to decide which twisted loop should be removed. After testing practical 3D models, the proposed algorithm is verified to use in tool path generation for 3D printing.