• Annual Conference of KIPS
• /
• 2023.11a
• /
• pp.271-273
• /
• 2023

불수의적인 혀 움직임과 떨림은 다른 신체 부위의 떨림증상이 없이 혀만 움직이는 증상을 말하며, 신경 정신과적 문제와 한의학의 기혈허약 등의 증상에 의해 발생할 수 있다. 혀 영상 촬영 장치로 정면과 측면의 혀를 연속을 촬영하여 혀의 움직임과 떨림을 탐색하려 한다. 혀의 표면은 코너와 같은 특징점을 구하기 어려운 모양이므로 혀의 움직임의 특성을 찾아내는 것은 매우 어려운 일이다. 움직임을 추적하는 방법 중에서 Farnebäck optical flow 방법은 모든 픽셀에 대해서 optical flow를 계산하여 혀의 움직임을 추적할 수 있었다. 이러한 움직임의 크기를 정면과 측면 영상에 대해서 구할 수 있었고, 움직임의 방향도 구할 수 있었다. 혀의 움직임과 떨림에 대한 위치별 정보와 세기 정보를 이용하여 건강상태를 진단할 수 있을 것으로 생각된다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.4 no.4
• /
• pp.15-22
• /
• 2003

A film scanner is one of the input devices for ac acquiring high resolution and high qualify of digital images from the existing optical film. Recently the demand of film scanners have risen for experts of image printing and editing fields. However, due to the nonlinear characteristic of light source and sensor, colors of the original film image do not correspond to the colors of the scanned image. Therefore color correction for the scanned digital image is essential in film scanner. In this paper, neural network method is applied for the color correction to CIE L/sup *//a/sup *//b/sup */ color model data converted from RGB color model data. Also a film scanner hardware with 12 bit color resolution for each R, G, B and 2400 dpi is implemented by using the TMS320C32 DSP chip and high resolution line sensor. An experimental result shows that the average color correction rate is 79.8%, which is an improvement of 43.5% than our previous method, polygonal regression method.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.44 no.2 s.314
• /
• pp.24-31
• /
• 2007

This paper presents self tuning PI(STPI) controller of IPMSM drive using neural network. In general, PI controller in computer numerically controlled machine process fixed gain. They may perform well under some operating conditions, but not all. To increase the robustness of fixed gain PI controller, STPI controller proposes a new method based neural network. STPI controller is developed to minimize overshoot, rise time and settling time following sudden parameter changes such as speed, load torque and inertia. Also, this paper is proposed speed control of IPMSM using neural network and estimation of speed using artificial neural network(ANN) controller. The back propagation neural network technique is used to provide a real time adaptive estimation of the motor speed. The results on a speed controller of IPMSM are presented to show the effectiveness of the proposed gain tuner. And this controller is better than the fixed gains one in terms of robustness, even under great variations of operating conditions and load disturbance.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.4
• /
• pp.1136-1142
• /
• 1999

A difficulty of the visual inspection for translated, magnified and rotated objects exists owing to the limitation of recognition rate. In this paper, we perform to define Integral Logarithm Transform(ILT), to consider its characteristic for implementation of Translation-, Magnification- and Rotation-invariant inspection system, and to compare with other methods in inspection error rate. By using magnification and rotation invariance properties of ILT, it makes easier than other methods to extract the rotation degree. The new method employs the ILT for the good/bad inspection of translated, magnified and rotated objects and experiment is performed to achieve translation, magnification and rotation invariance. In other methods both magnification and rotation invariance can't be available. As the result of he experiment, it is not better than the self-organizing map in the improvement of recognition rate, but it shows us the possibility to be used as a tool for the good/bad inspection system.

• KIISE Transactions on Computing Practices
• /
• v.23 no.2
• /
• pp.104-109
• /
• 2017

We propose a new machine learning algorithm which analyzes daily activity patterns of users from multi-modal wearable sensor data. The proposed model learns and extracts activity patterns using input from wearable devices in real-time. Inspired by cue integration of human's property, we constructed gated multi-modal neural networks which integrate wearable sensor input data selectively by using gate modules. For the experiments, sensory data were collected by using multiple wearable devices in restaurant situations. As an experimental result, we first show that the proposed model performs well in terms of prediction accuracy. Then, the possibility to construct a knowledge schema automatically by analyzing the activation patterns in the middle layer of our proposed model is explained.

• The KIPS Transactions:PartD
• /
• v.19D no.3
• /
• pp.237-246
• /
• 2012

Although the development of Global Positioning System (GPS) are more and more mature, its accuracy is just acceptable for outdoor positioning, not positioning for the indoor of building and the underpass. For the positioning application area for the indoor of building and the underpass, GPS even cannot achieve that accuracy because of the construction materials while the requirement for accurate positioning in the indoor of building and the underpass, because a space, a person is necessary, may be very small space with several square meters in the indoor of building and the underpass. The Received Signal Strength (RSS) based localization is becoming a good choice especially for the indoor of building and the underpass scenarios where the WiFi signals of IEEE 802.11, Wireless LAN, are available in almost every indoor of building and the underpass. The fundamental requirement of such localization system is to estimate location from Access Point (AP) to mobile device using RSS at a specific location. The Multi-path fading effects in this process make RSS to fluctuate unpredictably, causing uncertainty in localization. To deal with this problem, the combination for the method of Neural Networks and Push-Pull Estimation is applied so that the carried along the devices can learn and make the decision of position using mobile device where it is in the indoor of building and the underpass.

• Membrane Journal
• /
• v.12 no.3
• /
• pp.151-157
• /
• 2002

This research was to demonstrate the Possibility of sewage reuse for industrial purpose with use of membrane system. A bench scale test with microfiltration and reverse osmosis showed that microfiltration in the sewage treatment was not able to remove the soluble salts but 70% suspended solids (SS), suggesting that the treated water could be used as direct cooling water. In addition, the reverse osmosis removed not only soluble salts but also 95% SS, proposing that reverse osmosis-treated water could be used as both indirect cooling water and rinsing water. For a 100 ton/day pilot plant, 20 and 12 elements of microfiltration and reverse osmosis were required, respectively.

• Annual Conference on Human and Language Technology
• /
• 2000.10d
• /
• pp.405-410
• /
• 2000

본 연구에서는 기능적 자기공명영상을 이용하여 한글 문장의 의미와 통사 처리에 관한 뇌의 활성화 양상을 비교함으로써 한글문장 이해의 과정에 대한 신경해부학적 증거를 찾고자 하였다. 6명의 자원자를 대상으로 문장진위판단과제를 이용하여 활성화를 유도하였다. 1.5T 초전도 자기공명영상 장치에서 EPI로 BOLD 기법을 이용하여 기능적 영상을 얻었으며 영상 후 처리는 SPM99 분석 프로그램을 이용하였다. 의미관련 통사관련 문장 모두에서 좌 우 전두회(frontal gyrus) 영역에서 활성화되었다. 의미와 통사처리 영역을 구분하기 위하여 감산법을 적용한 결과, 의미처리는 좌반구의 중측두회(middle temporal gyrus) 영역에서, 통사처리는 우반구의 하전두회(BA44) 부위에서 더 많이 활성화되었다. 의미처리에서 더 우세한 성향을 띠는 부위로 밝혀진 중측두회 영역은 의미처리시에 활성화되는 영역으로 보고하는 기존의 연구와 일치하는 결과이다. 의미와 통사 문장처리시의 뇌 활성화 양상은 뇌의 여러 영역에서 중첩되어 있기는 하지만, 특정영역에서의 차이를 보이고 있으므로, 의미와 통사처리는 다른 기전(mechanism)에 의해서 일어남을 시사해 준다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.37 no.5
• /
• pp.76-83
• /
• 2000

This paper proposes a new color halftorning method using color correction by vector error diffusion to reduce color difference, necessarily appears on cross-media color reproduction In order to predict output colors on each device, a neural system IS applied and mean prediction errors in device characterization for monitor and printer are defined to calculate the thresholds for color correction Thus, color difference between monitor and printer is compared per each pixel If color difference is larger than the predetermined mean prediction errors, the halftoned dots to the current pixel are rearranged by vector error diffusion The proposed method can reduce the smear artifact by selective vector error diffusion and decrease color difference on cross- media color reproduction by color correction.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.6
• /
• pp.4328-4334
• /
• 2015

We determined a method to determine marine planktonic organism viability using Evan's blue, Aniline blue, and 5-choromethyfluorescein diacetate (CMFDA). The Evan's blue and Aniline blue methods produced bright blue light for dead phytoplankton and zooplankton and were the best dyes to detect dead cells. The staining efficiency of Evan's blue and Aniline blue were ${\geq}90%$ of the original field sample. However, it was difficult to test the efficiency of a ship's ballast water treatment system because detection of living cells. In contrast, the CMFDA method, which is based on measuring cell esterase activity using a fluorimetric stain, was the best dye to detect live cells of almost all phytoplankton species, and staining efficiency was 70%. The CMFDA method is similar to the fluorescein diacetate (FDA) staining method. Therefore, we estimated viability of phytoplankton species using a double-staining method by combining CMFDA and FDA to determine optimum staining efficiency. As a result, the frequency of dying cells based on the double-staining method was 95%, which was significantly higher than that of single CMDFA staining. Our results suggest that a CMDFA + FDA assay is more effective to determine survival of marine plankton and that this method was applicable to investigate the efficacy of a ship's ballast water treatment system.