• Proceedings of the KIEE Conference
• /
• 1993.11a
• /
• pp.91-93
• /
• 1993

This paper introduced an advanced PWM method to drive a variable speed AC motor. With this technique, a switching pattern is determined to minimize the error between a reference signal and feedback signal. In addition to its simplicity of implementation, the proposed technique has the advantage of control led constant voltage per frequency operation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.8
• /
• pp.1899-1907
• /
• 2013

This paper presents an efficient defect detection and classification system based on reference image for COF (Chip-on-Film) which encounters fatal defects after ultra fine pattern fabrication. These defects include typical ones such as open, mouse bite (near open), hard short and soft short. In order to detect these defects, conventionally it needs visual examination or electric circuits. However, these methods requires huge amount of time and money. In this paper, based on reference image, the proposed system detects fatal defect and efficiently classifies it to one of 4 types. The proposed system includes the preprocessing of the test image, the extraction of ROI, the analysis of local binary pattern and classification. Through simulations with lots of sample images, it is shown that the proposed system is very efficient in reducing huge amount of time and money for detecting the defects of ultra fine pattern COF.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.23 no.6
• /
• pp.440-443
• /
• 2010

We report a top-down approach based on atomic force microscopy (AFM) local anodic oxidation for the fabrication of the nano-pattern field effect transistors (FETs). AFM anodic oxidation is relatively a simple process in atmosphere at room temperature but it still can result in patterns with a high spatial resolution, and compatibility with conventional silicon CMOS process. In this work, we study nano-pattern FETs for various cross-bar distance value D, from ${\sim}0.5\;{\mu}m$ to $1\;{\mu}m$. We compare the optical characteristics of the patterned FETs and of the reference FETs based on both 2-dimensional simulation and experimental results for the wavelength from 100 nm to 900 nm. The simulated the drain current of the nano-patterned FETs shows significantly higher value incident the reference FETs from ${\sim}1.7\;{\times}\;10^{-6}A$ to ${\sim}2.3\;{\times}\;10^{-6}A$ in the infrared range. The fabricated surface texturing of photo-transistors may be applied for high-efficiency photovoltaic devices.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.1
• /
• pp.115-122
• /
• 1999

In this paper, we confirmed that a speech signal is a chaos signal, and in order to use it as a speaker recognition parameter, analyzed chaos dimension. In order to raise speaker identification and pattern recognition, by making up the strange attractor involving an individual's vocal tract characteristics very well and applying fuzzy membership function to correlation dimension, we proposed fuzzy correlation dimension. By estimating the correlation of the points making up an attractor are limited according space dimension value, fuzzy correlation dimension absorbed the variation of the reference pattern attractor and test pattern attractor. Concerning fuzzy correlation dimension, by estimating the distance according to the average value of discrimination error per each speaker and reference pattern, investigated the validity of speaker recognition parameter.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1183-1188
• /
• 2005

In this paper, we propose a 3D automated measuring system which measures the mandibular movements and the reference plane of the jaw movements. In diagnosis and treatment of the malocclusions, it is necessary to estimate the mandibular movements and the reference plane of the jaw movements. The proposed system is configured with double stereo-cameras, PC, two moving pattern plates(MPPs), two fixed pattern plates(FPPs) and one orbital marker. The virtual pattern plate is applied to calculate the homogeneous transformation matrices which describe the coordinates systems of the FPP and MPP with respect to the world coordinates system. To estimate the parameters of the hinge axis, the Euler's theorem is applied. The hinge axis points are intersections between the FPPs and the hinge axis. The coordinates of a hinge axis point with respect to the MPP coordinates system are set up to fixed value. And then, the paths of the jaw movement can be calculated by applying the homogeneous transformation matrix to fixed hinge axis point. To examine the accuracy of the measurements, experiments of measuring the hinge axis points and floating paths of them are performed using the jaw motion simulator. As results, the measurement errors of the hinge axis points are within reasonable boundary, and the floating paths are very similar to the simulator's moving path.

• Journal of the Korean Institute of Telematics and Electronics T
• /
• v.36T no.4
• /
• pp.30-37
• /
• 1999

This study suggested the Fuzzy Lyapunov dimension. The Fuzzy Lyapunov dimension is to evaluate the quantitative variation of the attractor. In this paper the speaker recognition is evaluated by the Fuzzy Lyapunov dimension. It has been proved that the suggested Fuzzy Lyapunov dimension is superior in the discrimination characteristics between standard reference pattern attractors, and in reference to the test pattern attractor, it has been verified that it is the speaker recognition parameter which absorbs the pattern variation. In order to evaluate the Fuzzy Lyapunov dimension as speaker recognition parameter, the mistaken recognition according to discrimination error in each of speaker and standard reference pattern was estimated, and the validity of the speaker recognition parameter was experimental. As the result of the speaker recognition experiment, 97.0[%] of recognition ratio was obtained, and it was confirmed that the Fuzzy Lyapunov dimension was fit for the speaker recognition parameter.

• Proceedings of the ESK Conference
• /
• 1996.10a
• /
• pp.197-205
• /
• 1996

Unification of neural network with a hierarchical pattern recognition is presented for recognizing large set of objects. A two-step identification procedure is developed for pattern recognition: coarse and fine identification. The coarse identification is designed for finding a class of object while the fine identification procedure is to identify a specific object. During the training phase a course neural network is trained for clustering larger set of reference objects into a number of groups. For training a fine neural network, expert neural network is also trained to identify a specific object within a group. The presented idea can be interpreted as two step identification. Experimental results are given to verify the proposed methodology.

• Horticultural Science & Technology
• /
• v.35 no.1
• /
• pp.108-120
• /
• 2017

Modern watermelon cultivars (Citrullus lanatus [Thunb.] Matsum.& Nakai var. lanatus) have fruits with diverse phenotypes, including fruit shape, rind patterns, and flesh color. Molecular markers enable efficient selection of plants harboring desirable phenotypes. In the present study, publicly available DNA markers tightly linked to fruit shape, rind stripe pattern, and flesh color were evaluated using 85 watermelon accessions with diverse fruit phenotypes. For fruit shape, the dCAPS SUN - Cla011257 marker revealed an 81% of marker - trait match for accessions with elongated or round fruits. For rind stripe pattern, the SCAR wsb6-11marker was effective for selecting Jubilee-type rind pattern from other rind patterns. For flesh color, the Clcyb.600 and Lcyb markers derived from a mutation in the Lycopene ${\beta}$ - cyclase (Lcyb) gene, were effective at selecting red or yellow flesh. Forty-eight accessions possessing diverse fruit - related traits were selected as a reference array and their genetic relationships assessed using 16 SSR markers. At a coefficient of 0.11, the 48 accessions grouped into two major clades: Clade I and Clade II. Clade I subdivided further into subclades I - 1 and I - 2 at a coefficient of 0.39. All accessions with colored flesh were classified into Clade I, whereas those with white - flesh were classified into Clade II. Differences in fruit traits between subclades I - 1 and I - 2 were observed for rind pattern and fruit color; a majority of the accessions with Crimson-type striped or non-striped rind were grouped together in subclade I - 1, while most accessions in subclade I - 2 had a Jubilee - type rind stripe pattern. These results imply that reference array watermelon accessions possess distinguishable genetic structure based on rind stripe pattern. However, no significant grouping pattern was observed based on other fruit-related traits.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.9
• /
• pp.133-141
• /
• 1994

Conventional competitive learning algorithms compute the Euclidien distance to determine the winner neuron out of all predetermined output neurons. In such cases, there is a drawback that the performence of the learning algorithm depends on the initial reference(=weight) vectors. In this paper, we propose a new competitive learning algorithm that dynamically generates output neurons. The proposed method generates output neurons by dynamically changing the class thresholds for all output neurons. We compute the similarity between the input vector and the reference vector of each output neuron generated. If the two are similar, the reference vector is adjusted to make it still more like the input vector. Otherwise, the input vector is designated as the reference vector of a new outputneuron. Since the reference vectors of output neurons are dynamically assigned according to input pattern distribution, the proposed method gets around the phenomenon that learning is early determined due to redundant output neurons. Experiments using speech data have shown the proposed method to be superior to existint methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.999-1004
• /
• 1989

This paper discusses a new method of recognizing patterns employing consistent labeling. A consistent labeling problem is a generalized expression of constraint satisfaction problems. When a pattern is recognized by pattern matching, the matching between a reference pattern and an acquired pattern resolves itself into finding correspondence between the pixels on the former and those on the latter. This can be expressed as a consistent labeling problem. Pattern association, a variation of pattern recognition, is also described employing consistent labeling. The proposed technique is supported by experimental results, yet further studies need to be done before its practical use.