• Journal of Multimedia Information System
• /
• v.4 no.2
• /
• pp.79-82
• /
• 2017

Smoke and fire have different shapes and colours. This article suggests a fully connected system which is used two features using Adaboost algorithm for constructing a strong classifier as linear combination. We calculate the local histogram feature by gradient and bin, local binary pattern value, and projection vectors for each cell. According to the histogram magnitude, this paper applied adapted weighting value to improve the recognition rate. To preserve the local region and shape feature which has edge intensity, this paper processed the normalization sequence. For the extracted features, this paper Adaboost algorithm which makes strong classification to classify the objects. Our smoke detection system based on the proposed approach leads to higher detection accuracy than other system.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.1E
• /
• pp.27-31
• /
• 1999

Recent linguistic researches have brought into focus the relations between prosodic structures and syntactic, semantic or phonological structures. Most of them prove that prosodic information is available for understanding syntactic, semantic and discourse structures. But this result has not been integrated yet into recent Korean speech recognition or understanding systems. This study, as a part of integrating prosodic information into the speech recognition system, proposes an automatic detection technique of Korean accentual phrase boundaries by using one-stage DP, and the normalized pitch pattern. For making the normalized pitch pattern, this study proposes a method of modified normalization for Korean spoken language. For the experiment, this study employs 192 sentential speech data of 12 men's voice spoken in standard Korean, in which 720 accentual phrases are included, and 74.4% of the accentual phrase boundaries are correctly detected while 14.7% are the false detection rate.

• Water for future
• /
• v.5 no.2
• /
• pp.30-43
• /
• 1972

The paper describes on the hydrologic analysis of point rainfall data of the three major areas, such as in Seoul, Pusan and Taegu. Scheme of the paper is analyzed five research cases. Contents of the analysis are carried out five kinds of transformed variables for determination of rainfall distribution types and two kinds of reliability tests on unusual(extraordinary) values each rainfall durations:short durations, long durations, long durations, monthly and yearly. Rainfall depth probability had been computed methods of hydrologic amounts analysis namely logarithmic transformations or Gumbel-Chow method and so on, but in this paper it is calculated log xi, n-square root transformations by using normal distribution function and normalization of rainfall distributions is examined graphical tests and $X^2-test$(chi-square test). Furthermore, rainfall depth probability is calculated taking into account the safty factors of project life of hydraulic structures. We think it is advanced contents that considering priceless experiences, the life of structures, conditions and more problems of planning engineers and designers, proposed rainfall amounts(proposed values) are presented charts or figures.

• The Journal of Korea Robotics Society
• /
• v.12 no.4
• /
• pp.432-440
• /
• 2017

In this paper, we propose a jellyfish distribution recognition and monitoring system using a UAV (unmanned aerial vehicle). The UAV was designed to satisfy the requirements for flight in ocean environment. The target jellyfish, Aurelia aurita, is recognized through convolutional neural network and its distribution is calculated. The modified deep neural network architecture has been developed to have reliable recognition accuracy and fast operation speed. Recognition speed is about 400 times faster than GoogLeNet by using a lightweight network architecture. We also introduce the method for selecting candidates to be used as inputs to the proposed network. The recognition accuracy of the jellyfish is improved by removing the probability value of the meaningless class among the probability vectors of the evaluated input image and re-evaluating it by normalization. The jellyfish distribution is calculated based on the unit jellyfish image recognized. The distribution level is defined by using the novelty concept of the distribution map buffer.

• Elastomers and Composites
• /
• v.53 no.2
• /
• pp.86-94
• /
• 2018

The effect of the NCO index and catalyst type on the thermal stability of poly(urethane-isocyanurate) (PUIR) foams was investigated to identify a method for enhancing the flame resistance of PUIR. PUIR foams were prepared using 4,4-diphenylmethane diisocyanate (MDI) and [(diethylene glycol)adipate]diol, which were synthesized by esterification of adipic acid and diethylene glycol. Dabco K-15, Dabco TMR-30, and Toyocat RX-5 were used as the catalysts for trimerization and gelation. The amount of urea and isocyanurate groups in PUIR was semi-quantitatively determined by normalizing their absorbance with the phenyl absorbance measured by FT-IR. The normalization data showed that Dabco TMR-30 effectively generated isocyanurate groups in PUIR. As a result, Dabco TMR-30 effectively raised the decomposition temperature and increased the 800 K and 900 K residues of the PUIR foam synthesized with an NCO index of 200.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2002.10a
• /
• pp.245-252
• /
• 2002

The roughness of Joomoonjin sand and the Dongchun river sand was analysed by the fractal theory. It was found that the fractal dimension(D$\_$F/) of Joomoonjin sand is a little smaller than the Dongchun river sand. That means Joomoonjin sand is smoother than the Dongchun river sand. The measurements of D$\_$F/ of different fraction of the Donchun river sand showed that large particles were rougher than fine particles. The shapes of both sands were analysed by the Discrete Fourier Transform(DFT) and the Grid-based(GB) method. Normalization of coefficients with respect to size, starting point and its orientation made the coefficients invaried to these characteristics. The mean of the normalized coefficients was used to reconstruct the average shape for both sands, respectively. The measurements of the ellipticity ratio of different fraction of both sands showed that Joomoonjin sand is slightly flatter than the Dongcun river sand.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 1999.05a
• /
• pp.161-164
• /
• 1999

Estimation system of aging diagnosis using partial discharge(PD) is being highlighted as a research area for the residual lifetime pridiction of industrial equipment. But the application of PD requires complicated analysis method as expert system because the PD has complex progressing forms according to external stress. In this paper, it has been investigated the statistical distribution to express the 2D PD patterns of the diagnosis system using neural network(NN).

• KSBB Journal
• /
• v.16 no.4
• /
• pp.381-388
• /
• 2001

A program for integrated gene expression profile analysis such as hierarchical clustering, K-means, fuzzy c-means, self-organizing map(SOM), principal component analysis(PCA), and singular value decomposition(SVD) was made for DNA chip data anlysis by using Matlab. It also contained the normalization method of gene expression input data. The integrated data anlysis program could be effectively used in DNA chip data analysis and help researchers to get more comprehensive analysis view on gene expression data of their own.

• Geomechanics and Engineering
• /
• v.12 no.2
• /
• pp.269-294
• /
• 2017

This study aimed to realize the creation of fuzzy stochastic damage to describe reliability more essentially with the analysis of harmony of damage conception, probability and fuzzy degree of membership in interval [0,1]. Two kinds of fuzzy behaviors of damage development were deduced. Fuzzy stochastic damage models were established based on the fuzzy memberships functional and equivalent normalization theory. Fuzzy stochastic damage finite element method was developed as the approach to reliability simulation. The three-dimensional fuzzy stochastic damage mechanical behaviors of Jianshan mine slope were analyzed and examined based on this approach. The comprehensive results, including the displacement, stress, damage and their stochastic characteristics, indicate consistently that the failure foci of Jianshan mine slope are the slope-cutting areas where, with the maximal failure probability 40%, the hazardous Domino effects will motivate the neighboring rock bodies' sliding activities.

• Journal of Institute of Control, Robotics and Systems
• /
• v.3 no.6
• /
• pp.588-593
• /
• 1997

This paper shows a design and implementation of throttle valve controller for engine dynamometer system using fuzzy logic. Recently, we demanded the excellent measuring equipment so as to improve engine performance. The throttle valve control for engine dynamometer system is a very particular part in the engine control. Since the structure of engine dynamometer system is very complicated and has nonlinear elements which are influenced by disturbance of vibration, heating, cooling, and energy loss so on. In this paper, fuzzy logic control application have been successful in throttle valve control problem for engine dynamometer system in which the conventional control had difficulties dealing with the system. In this study, we propose a method that the control strategy uses Fuzzy Look-up table and normalization and obtained the satisfying result from realized throttle valve controller for engine dynamometer system.