• Journal of Electrical Engineering and information Science
• /
• v.3 no.3
• /
• pp.286-291
• /
• 1998

A remote measurement system for nerve conduction has been developed to aid patients with spinal cord injury by accident. Existing cooperation between rescuers and doctors can be supported through the introduction of multimedia desktop video conferencing. Such facilities provide several advantages over conventional video conferencing. In particular, patients may feel free because they can see a doctor through the video conferencing facilities. This paper describes the system implementation and evaluation. The author considers the network capability and image data handling, and introduces a method a transmit image data for this system.

• Nuclear Engineering and Technology
• /
• v.53 no.1
• /
• pp.294-303
• /
• 2021

Various types of sensors are used at industrial sites to measure vibration. With the increase in the diversity of vibration measurement methods, vibration monitoring methods using camera equipment have recently been introduced. However, owing to the physical limitations of the hardware, the measurement resolution is lower than that of conventional sensors, and real-time processing is difficult because of extensive image processing. As a result, most such methods in practice only monitor status trends. To address these disadvantages, a high-resolution vibration measurement method using image analysis of the edge region of the structure has been reported. While this method exhibits higher resolution than the existing vibration measurement technique using a camera, it requires significant amount of computation. In this study, a method is proposed for rapidly processing considerable amount of image data acquired from vision equipment, and measuring the vibration of structures with high resolution. The method is then verified through experiments. It was shown that the proposed method can fast measure vibrations of structures remotely.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.2
• /
• pp.165-170
• /
• 2020

The spiking neural network operates in a different mechanism than the existing neural network. The existing neural network transfers the output value to the next neuron via an activation function that does not take into account the biological mechanism for the input value to the neuron that makes up the neural network. In addition, there have been good results using deep structures such as VGGNet, ResNet, SSD and YOLO. spiking neural networks, on the other hand, operate more like the biological mechanism of real neurons than the existing activation function, but studies of deep structures using spiking neurons have not been actively conducted compared to in-depth neural networks using conventional neurons. This paper proposes the method of loading an deep neural network model made from existing neurons into a conversion tool and converting it into a spiking deep neural network through the method of replacing an existing neuron with a spiking neuron.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.77-84
• /
• 2022

In this paper, we propose a novel unsupervised feature selection method. Conventional unsupervised feature selection method defines virtual label and uses a regression analysis that projects the given data to this label. However, since virtual labels are generated from data, they can be formed similarly in the space. Thus, in the conventional method, the features can be selected in only restricted space. To solve this problem, in this paper, features are selected using orthogonal projections and low-rank approximations. To solve this problem, in this paper, a virtual label is projected to orthogonal space and the given data set is also projected to this space. Through this process, effective features can be selected. In addition, projection matrix is restricted low-rank to allow more effective features to be selected in low-dimensional space. To achieve these objectives, a cost function is designed and an efficient optimization method is proposed. Experimental results for six data sets demonstrate that the proposed method outperforms existing conventional unsupervised feature selection methods in most cases.

• Proceedings of the KIEE Conference
• /
• 2005.10b
• /
• pp.149-151
• /
• 2005

In this paper, we propose the Optimal Polynomial Neural Networks(PNN) for nonlinear process. The PNN is based on Group Method of Data Handling(GMDH) method and its structure is similar to feedforward Neural Networks. But the structure of PNN is not fixed like in conventional Neural Networks and can be generated. The each node of PNN structure uses several types of high-order polynomial such as linear, quadratic and modified quadratic, and is connected as various kinds of multi-variable inputs. The conventional PNN depends on experience of a designer that select No. of input variable, input variable and polynomial type. Therefore it is very difficult a organizing of optimized network. The proposed algorithm identified and selected No. of input variable, input variable and polynomial type by using Genetic Algorithms(GAs). In the sequel the proposed model shows not only superior results to the existing models, but also pliability in organizing of optimal network. Medical Imaging System(MIS) data is simulated in order to confirm the efficiency and feasibility of the proposed approach in this paper.

• Journal of Korea Water Resources Association
• /
• v.34 no.5
• /
• pp.567-575
• /
• 2001

A methodology is developed for designing the minimum-cost water distribution network. The method is based on network simulations and an optimization scheme using genetic algorithms. Being a stochastic optimization scheme, genetic algorithms have advantages over the conventional search algorithms in solving network problems known for their nonlinearities and herculean computational costs. While existing methods focus on the design of either entirely new or parallel augmentation of network systems, the proposed method can be applied to problems having both new branches of tree-type and paralle augmentation in loops. The applicability of the method was shown through a case study for Baekryeon water supply system. The optimized design resulted in the maximum 5.37% savings compared to the conventional design without optimization, while meeting the hydraulic constraints.

• Journal of Korea Multimedia Society
• /
• v.10 no.10
• /
• pp.1229-1239
• /
• 2007

In conversion of web pages for mobile devices, since existing web conversion systems eliminate unnecessary objects indiscriminately or only convert the layout of a page, the original intention of a web designer is spoiled. In spite of great advances in device technology, there still remain the problems of limited display sire and insufficient CPU power. In this paper, we propose an adaptation method that minimizes the loss of information and preserves the original intention of a web designer from the visual aspect. In image conversion, the proposed method extracts meaningful information from the background against conventional resizing methods, and performs layout-based conversion for the arrangement of image components. Experimental results show the efficiency of the method on the accuracy and the usability in comparison with existing methods.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.19 no.1
• /
• pp.80-88
• /
• 2010

In a conventional system, hydraulic unit pumps out pressurized oil when the power use is not even necessary. As a result, it causes much power loss. This study is on the proposal of new hydraulic unit which controls the revolution of the pump in order to produce proper power needed and to have good response characteristic. In addition, the existing control methods such as PID control method, fuzzy control method, and adaptive control method are applied to the proposed hydraulic unit. Then the best control method is selected and the controller is developed to realize minimum power loss.

• Transactions of Materials Processing
• /
• v.9 no.3
• /
• pp.274-283
• /
• 2000

A dynamic explicit finite element code for simulating sheet forming processes has been developed. The code utilizes the discrete Kirchhoff shell element and contact force is treated by a conventional penalty method. In order to reduce the computational cost, a new and robust contact searching algorithm has been developed and implemented into the code. In the method, a hierarchical structure of tool segments is built for each tool at the initial stage of the analysis. hierarchical structure is built in a way to divide a box to 8 sub-boxes, 2 in each direction, until the lowest level of the hierarchical structure contains exactly one segment of the tool or empty. Then at each time step, contact is checked from the box to sub-boxes hierarchically for each node. Comparisons of computational results of various examples with the existing ones show the validity of the method.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.17 no.4
• /
• pp.337-344
• /
• 2012

A matrix converter is used for converting AC/AC power directly. In order to generate sinusoidal input/output waveform in matrix converter, it uses nine bidirectional switches and PWM modulation. This paper presents an analytical averaged loss model of matrix converter with DDPWM(direct duty ratio PWM) and proposes a new switching method for reducing switching losses. A Mathematical loss models with average magnitude of voltage/current are classed as conduction and switching loss. The proposed switching pattern is improved with existing DDPWM. To prove improved efficiency with proposed DDPWM, this paper compares losses between suggested switching pattern and conventional switching pattern using mathematical and simulation method. Each loss types are influenced by environmental factors such as temperature, switching frequency, output current and modulation method.