• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.05a
• /
• pp.1074-1079
• /
• 2007

We will introduce speech/speaker recognition algorithm for the isolated word. In general case of speaker verification, Gaussian Mixture Model (GMM) is used to model the feature vectors of reference speech signals. On the other hand, Dynamic Time Warping (DTW) based template matching technique was proposed for the isolated word recognition in several years ago. We combine these two different concepts in a single method and then implement in a real time speaker/speech recognition system. Using our proposed method, it is guaranteed that a small number of reference speeches (5 or 6 times training) are enough to make reference model to satisfy 90% of recognition performance.

• Journal of the Korean Society of Combustion
• /
• v.19 no.2
• /
• pp.28-36
• /
• 2014

Bio fuel oil combustion experiments were successfully demonstrated at the 75 MWe oil-fired power plant without major equipment retrofit and 100% bio-fuel oil combustion was possible without big problems. The experimental data error correction was conducted and numerical model-based analysis technique was applied for the evaluation of the results. Incase of bio fuel oil combustion, heat absorption of radiative heat transfer section was reduced while convection section has opposite trend. The furnace exit gas temperature tends to rise slightly. Environment emissions such as NOx and SOx concentrations showed a tendency to decrease during the bio fuel oil combustion period. On the other hand, boiler efficiency was slightly underestimated.

• The Journal of the Korean life insurance medical association
• /
• v.2 no.1
• /
• pp.3-16
• /
• 1985

Necessity of life insurance is stronger for people who feel some anxiety of their health. However, in fact, it is not permitted for them to get a contract, because life insurance stands on the mutual benefit system. Life insurance must be impartial to all applicants. However, it is very reasonable that an applicant, who has high medical impairment like heart infarction or cancer, is rejected, to have a contract by underwriting decision. On the other hand, if his medical impairment is not so severe, we might accept his application by giving some restriction. Numericalratingsystem by hunter-rogers gave us one of solutions to this problem. We can keep impartiality by using more restrictive decision, in order that we demand additional payment to the impairment applicant according to his mortality. We call this system as substandard life insurance. In this system we need detail information about impairments of applicants in order to decide the condition of substandard risks. Therefore, medical examiners are required to have high diagnostic technique.

• The Korean Journal of Pain
• /
• v.11 no.1
• /
• pp.138-142
• /
• 1998

Fibromyalgia syndrome is defined as a diffuse, aching musculoskeletal pain associated with multiple and discrete predictable tender points along with stiffness. A primary form associated without any disease is uncommon compared to the secondary diffuse fibrositis. We witnessed a case of a primary fibromyalgia syndrome. A 28-year-old woman was suffering from pain on posterior neck, right shoulder, upper back, left hand and arm for two years. She also complained of morning stiffness, fatigue, and headache. We treated her with stellate ganglion block with 0.25% bupivacaine, medication that included amitriptyline, TENS and superlizer. We also recommended aerobic exercise.

• 연구논문집
• /
• s.23
• /
• pp.155-164
• /
• 1993

A drawing technique which use only fine curves and dashed lines is widely adopted in banknotes and post stamps printing. Engraving of the curves and lines are traditionally performed by hand skill which provides low productivity in printing. As an effort for higher productivity and quality, a drawing automation method which can easily produce NC codes and drawings for a mechanical or chemical engraving is proposed. An initial work shows that it is possible to draw a portrait by contolling the width and length of predetermined fine lines according to the gray scale at the end points of each line. User interface functions of a commercial CAD system are heavily employed to exploit the presented method.

• Proceedings of the KIEE Conference
• /
• 1987.07b
• /
• pp.1394-1397
• /
• 1987

It is difficult to detect edge segments from a noisy image since the image have a noise in piratical applications which utilize some type of visual input capability. Hence, the proposed algorithm consists of the modality tests based on parallel statistical tests without a noise removal preprocessing or postprocessing, and the edge detection technique With one-Pixel edge segments in this paper. The algorithm is very reliable and effective in the case of those situations where the Picture is poor quality and low resolution. And it does'nt require thinning operation and thresholding in hand. Experimental comparision With the more conventional techniques when applied to typical low-quality Pictures confirms good capabilities of the algorithm.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.51 no.3
• /
• pp.100-107
• /
• 2002

In this paper, we propose an expert system for electronic interlocking which enhances the safty, efficiency and expanability of the existing system by designing real-time interlocking control based on the interlocking table automatically generated using artificial intelligence approach. The expert system consists of two parts; an interlocking table generation part and a real-time interlocking control part. The former generates automatically the interlocking relationship of all possible routes by searching dynamically the station topology which is obtained from station database. On the other hand, the latter controls the status of station facilities in real-time by applying the generated interlocking relationship to the signal facilities such as signal devices, points, track circuits for a given route. The expert system is implemented in C language which is suitable to implement the interlocking table generation part using the dynamic memory allocation technique. Finally, the effectiveness of the expert system is proved by simulating for the typical station model.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.41 no.2
• /
• pp.107-117
• /
• 1992

This paper presents an improved algorithm of optimal measurement design with a reliability evaluation method for a large power system. The proposed algorithm is developed to consider the dummy bus and to achieve highest accuracy of the state estimator as well with the limited investment cost. The dummy bus in the power system is impossible to install measurement meter, while real and reactive power measurements is considered in the proposed algorithm. On the other hand, P/C model is developed by taking advantage of the matrix sparsity. The improved program is successfully tested for KEPCO system with PSS/E lineflow calculated data package.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.713-716
• /
• 2006

The technique of the information communication are advanced recently and a performance enhance of a hand carried computing device was developed rapidly. Mobile Communication Carrier developed currently the phone navigation and are carrying out the service. But such service localizes at the vehicle movement. In This paper, we explain a system structure for the pedestrian navigation of the Wireless Internet Platform for Interoperability(WIPI) mobile phone which contains the MS-Based Global Positioning System(GPS) internally. And we verified the result to be developed by this method that proposes.

• Proceedings of the Korean Statistical Society Conference
• /
• 2003.10a
• /
• pp.69-72
• /
• 2003

The paper considers partial least squares (PLS) as a new dimension reduction technique for the feature vector to overcome the small sample size problem in face recognition. Principal component analysis (PCA), a conventional dimension reduction method, selects the components with maximum variability, irrespective of the class information. So PCA does not necessarily extract features that are important for the discrimination of classes. PLS, on the other hand, constructs the components so that the correlation between the class variable and themselves is maximized. Therefore PLS components are more predictive than PCA components in classification. The experimental results on Manchester and ORL databases show that PLS is to be preferred over PCA when classification is the goal and dimension reduction is needed.