• The Journal of the Acoustical Society of Korea
• /
• v.13 no.3
• /
• pp.25-32
• /
• 1994

This paper presents the Korean digit recognition method under noise environment using the spectral mapping training based on static supervised adaptation algorithm. In the presented recognition method, as a result of spectral mapping from one space of noisy speech spectrum to another space of speech spectrum without noise, spectral distortion of noisy speech is improved, and the recognition rate is higher than that of the conventional method using VQ (vector quatization) and DTW(dynamic time warping) without noise processing, and even when SNR level is 0dB, the recognition rate is 10 times of that using the conventional method. It has been confirmed that the spectral mapping training has an ability to improve the recognition performance for speech in noise environment.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.35 no.9
• /
• pp.783-791
• /
• 2007

For an access to specified helicopter simulator qualification's level C or FTD(Flight Training Device) level 5 of FAA(Federal Aviation Administration) AC(Advisory Circular) 120-63, the mathematical model of a single rotor helicopter flight dynamics is investigated. From the rotorcraft simulation model validated by evaluation of its flight performance, the feasibility of the flight dynamic model that is selected for its effectiveness has been proved. Thereby the simulation environment for evaluation of helicopter flying quality is established with the development of FTD for training and testing the flight performance.

• International Journal of Advanced Culture Technology
• /
• v.8 no.2
• /
• pp.58-67
• /
• 2020

In this paper, the maintenance training manual of EV vehicle was produced by utilizing digital twin technology and various sensors such as IR-based light house tracking and head tracker. In addition, through digital twin technology and VR to provide high immersiveness to users, sensory content creation technology was secured through animation and effect realization suitable for EV vehicle maintenance situation. EV vehicle maintenance training manual is 3D engine programming and real-time creation of 3D objects and minimization of screen obstacles and selection of specific menus in virtual space in the form of training simulation. In addition, automatic output from the Head Mount Display (HUD), EV vehicle maintenance and inspection, etc., user can easily operate content was produced. This technology development can enhance immersion to users through implementation of detailed scenarios for maintenance / inspection of EV vehicles" and 3D parts display by procedure, realization of animations and effects for maintenance situations. Through this study, familiarity with improving the quality of education and safety accidents and correct maintenance process and the experienced person was very helpful in learning how to use equipment naturally and how to maintain EV vehicles.

• Journal of Korean Neurosurgical Society
• /
• v.57 no.1
• /
• pp.65-67
• /
• 2015

Cavernous hemangiomas were first reported in 1929 by Globus and Doshay, and are defined as benign vascular structures developed between the neural tissues occurring in the central nervous system, consisting of a dilated vascular bed. Cavernous hemangiomas comprise nearly 5-12% of all spinal vascular malformations; however, existence in the epidural space without bone involvement is rare. Only 4% of all cavernous hemangiomas (0.22/1.000.000) are purely epidural cavernous hemangiomas. In this case report, we removed a hemorrhagic thoracic mass presenting with progressive neurological deficits in a 55-year-old male patient. We found this case to be appropriate for presentation due to the rare occurrence of this type of cavernous hemangioma.

• Journal of information and communication convergence engineering
• /
• v.8 no.3
• /
• pp.267-272
• /
• 2010

For the last decade, recurrent neural networks (RNNs) have been commonly applied to communications channel equalization. The major problems of gradient-based learning techniques, employed to train recurrent neural networks are slow convergence rates and long training sequences. In high-speed communications system, short training symbols and fast convergence speed are essentially required. In this paper, the derivative-free Kalman filter, so called the unscented Kalman filter (UKF), for training a fully connected RNN is presented in a state-space formulation of the system. The main features of the proposed recurrent neural equalizer are fast convergence speed and good performance using relatively short training symbols without the derivative computation. Through experiments of nonlinear channel equalization, the performance of the RNN with a derivative-free Kalman filter is evaluated.

• Korean Institute of Interior Design Journal
• /
• no.33
• /
• pp.90-98
• /
• 2002

The purpose of this study is total analysis which is satisfied factors to users to object of inservice training institute in large enterprise. Common factors are inservice training building, outer space and unit of abode. Those are deduced in the planing stage by important degree. Methods of statistical analysis are enforced by analysis of primary factor, correlation and multiple regression. The result of this shows a difference between important orders to users according to character of influenced factors of the inservice training building and abode.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.12
• /
• pp.2613-2625
• /
• 1997

Theoretically, a multi-layered feedforward network has been known to be able to approximate a continuous function to an arbitrary degree of accuracy. However, these networks fail to approximate discontinuous functions when they are trained by well-known training algorithms. This paper presents a training algorithm which doesn't work consists of one or more modules, which are trained in a sequential order within subspaces of the input space, and is trained very rapidely once all modules are trained and merged. The experimantal results of applying this method indicates the proposed training algorithm is superior to traditional ones such as baskpagation.

• Proceedings of the KIEE Conference
• /
• 2004.11b
• /
• pp.91-93
• /
• 2004

Load forecasting is essential in order to supply electrical energy stably and economically in power systems. ANNs have flexibility to predict a nonlinear feature of load profiles. In this paper, we selected just the necessary input variables used in the paper(2) which is based on the phase-space embedding of a load time-series and reviewing others. So only 5 input variables were selected to forecast for spring, fall and winter season and another input considering temperature sensitivity is added during the summer season. The training cases are also selected from all previous data composed training cases of a 7-day, 14-day and 30-day period. Finally, we selected the training case of a 7-day period because it can be used in STLF without sacrificing the accuracy of the forecast. This allows more compact ANNs, smaller training cases. Consequently, test results show that compact neural networks can be forecasted without sacrificing the accuracy.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.3
• /
• pp.463-470
• /
• 2023

In this study, an electronic parts assembly training system using Leap Motion was developed in consideration of the processes actually operated in the assembly process of electronic products. Based on Leap Motion and Oculus VR equipment, the system was developed to transfer user's hand movement data in real time and convert it into hand movement in virtual space so that electronic parts assembly simulation can be performed step by step. Through this, it was confirmed that the user can obtain an experience similar to the actual electronic parts assembly work, prevent errors that may occur during the assembly process, and improve proficiency. It is expected that this thesis will provide directions for the quality improvement and development of various education and training programs for virtual reality-based manufacturing processes.

• International journal of advanced smart convergence
• /
• v.9 no.4
• /
• pp.52-61
• /
• 2020

The AHP method was used in 30 provinces of China to construct the index system of creative space efficiency evaluation and determine the weight of each index. The fuzzy comprehensive evaluation method was further used to score the indexes at all levels, and then the total efficiency score was sorted. The purpose of this study is to adjust the regional layout of creative space reasonably and implement financial policies accurately through the evaluation of the efficiency of creative space. The results is ranking top in weight of several indicators, which include the number of incubated Startups, the number of innovation and entrepreneurship mentors, the survival rate of incubator, the innovative training activities, etc. It was also found that Beijing, Shanghai, Jiangsu, Guangdong and Zhejiang ranked first in the score of creative space efficiency. This study is meaningful in that it was In order to effectively solve the problem of the imbalance of the creative space efficiency in China's province, by coordinating the regional pattern, establishing a sound service system and improving the efficiency evaluation system.