• The Journal of the Acoustical Society of Korea
• /
• v.14 no.1
• /
• pp.13-21
• /
• 1995

In this paper, we improve an image classifier algorithm based on neural network learning. It consists of two steps. The first is input pattern generation and the second, the global neural network implementation using an improved back-propagation algorithm. The feature vector for pattern recognition consists of the codebook data obtained from self-organization feature map learning. It decreases the input neuron number as well as the computational cost. The global neural network algorithm which is used in classifier inserts a control part and an address memory part to the back-propagation algorithm to control weights and unit-offsets. The simulation results show that it does not fall into the local minima and can implement easily the large-scale neural network. And it decreases largely the learning time.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.2
• /
• pp.407-412
• /
• 2009

Recently, in Weld Processing field, unmanned and automatic system construction has experienced the rapid growth, and diverse signal processing has been employed in order to translate the exact weld pattern. In this paper, We will suggest the effective neural network which can decide the weld quality in arc weld and monitoring system in real time. In addition, We will present the pre-processing for selecting the study data, and the method to evaluate the wave of weld more precisely and accurately through known Neural Network.

• BMB Reports
• /
• v.41 no.1
• /
• pp.23-28
• /
• 2008

The altered amino acid (AA) levels as neurotransmitter closely correlate to neurodegenerative conditions including Alzheimer's disease (AD). Target profiling analysis of nineteen AAs in brain cortex samples from three Tg2576 mice as AD model and three littermate mice as control model was achieved as their N(O,S)-ethoxycarbonyl/tert-butyldimethylsilyl derivatives by gas chromatography. Subsequently, star pattern recognition analysis was performed on the brain AA levels of AD mice after normalization to the corresponding control median values. As compared to control mice, $\gamma$-aminobutyric acid among ten AAs found in brain samples was significantly reduced (P < 0.01) while leucine was significantly elevated (P < 0.02) in AD mice. The normalized AA levels of the three AD mice were transformed into distorted star patterns which was different from the decagonal shape of control median. The present method allowed visual discrimination of the three AD mice from the controls based on the ten normalized AA levels.

• Natural Product Sciences
• /
• v.25 no.4
• /
• pp.334-340
• /
• 2019

Cuscuta chinensis Lam. and Cuscuta japonica Choisy are parasitic plants. C. chinensis seeds were traditionally used for treatment of kidney and liver deficiencies. C. japonica seeds were used as tonic medicine to improve liver function and strengthen kidneys, treatment of high blood pressure, chronic diarrhea, and sore eyes. Cuscutae Semen are seeds of only C. chinensis in Korean Herbal Pharmacopoeia (K.H.P.). The developed HPLC-PDA method easily, accurately, and sensitively quantified using eight marker compounds [hyperoside (1), astragalin, (2), quercetin (3), kaempferol (4), chlorogenic acid (5), 3,4-di-O-caffeoylquinic acid (6), 1,5-di-O-caffeoylquinic acid (7), and 4,5-di-O-caffeoylquinic acid (8)]. In addition, the method may be used to distinguish seeds between C. chinensis Lam. and C. japonica Choisy. Furthermore, the result from the current study was applied to clarify samples between steam processed and unprocessed samples of C. chinensis by pattern analysis.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.1
• /
• pp.23-30
• /
• 2014

Feature matching using image descriptors is robust method used recently. However, mismatches occur in 3D transformed images, illumination-changed images and repetitive-pattern images. In this paper, we observe that there are a lot of mismatches in the images which have repetitive patterns. We analyze it and propose a method to eliminate these mismatches. MDMF(Multiple Descriptors-based Mismatch Filtering) eliminates mismatches by using descriptors of nearest several features of one specific feature point. In experiments, for geometrical transformation like scale, rotation, affine, we compare the match ratio among SIFT, ASIFT and MDMF, and we show that MDMF can eliminate mismatches successfully.

• Journal of the Korea Furniture Society
• /
• v.24 no.4
• /
• pp.400-407
• /
• 2013

21C modern society is a time when enough design to dominate the world, a unique design is required new in all aspects. In addition, there is a religion to the best of Zen art. In our cultural heritage with a ethnicity unique unparalleled in the world, and a new recognition of traditional culture and identity of its own in our country, lattice door flower temple the beauty of Korea in the way you tell, it was modernized by interpreting a new one to utilize as a design element of a wood lattice door flower temple. In this study, to re-appear in openwork having regard to the symbolism of pattern and shape of the lattice door flower temple molding properties were highlighted, change the part of the motif, repeated the curve geometric has been designed to simplify as. As a result, it is possible to present the possibility of a design element that can design a traditional Korean is reinterpreted Modern thereby simplify the repetitive element of the related art, for a new aesthetic changes unique. Design Development of wood with a Buddhist temple flower lattice modern re-interpretation, it suits while a strong indication Korea imagery and expressed a variety of needs and personality to the design elements that are competitive in the world the cage, and design of wood, various studies competitive on to be applied to each field is required.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2000.05a
• /
• pp.829-832
• /
• 2000

This paper describes a research work of developing a computer-aided design of product with bending and piercing for progressive working. An approach to the CAD system is based on the knowledge-based rules. Knowledge fur the CAD system is formulated from plasticity theories, experimental results and the empirical knowledge of field experts. The system has been written in AutoLISP on the AutoCAD with a personal computer and is composed of three main modules, which are input and shape treatment, flat pattern layout and strip layout module. Based on knowledge-based rules, the system is designed by considering several factors, such as radius and angle of bend, material and thickness of product, complexities of blank geometry and punch profile, bending sequence, and availability of press. Strip layout drawing automatically generated by piercing with punch profiles divided into for external area is simulated in 3-D graphic forms, including bending sequences for the product with piercing and bending. Results obtained using the modules enable the manufacturer of electronic products to be more efficient in this field.

• Atmosphere
• /
• v.21 no.3
• /
• pp.221-228
• /
• 2011

The nonlinear characteristics of summer monsoon intraseasonal oscillation (ISO) in precipitation, which is manifested as fluctuations in convection and circulation, is one of the major difficulty on the prediction of East Asian summer monsoon (EASM). The present study aims to identify the spatial distribution and time evolution of nonlinear phases of monsoon ISO. In order to classify the different phases of monsoon ISO, Self-Organizing Map(SOM) known as a nonlinear pattern recognition technique is used. SOM has a great attractiveness detecting self-similarity among data elements by grouping and clustering such self-similar components. The four important patterns are demonstrated as Meiyu-Baiu, Changma, post-Changma, and dry-spell modes. It is found that SOM well captured the formation of East Asian monsoon trough during early summer and its northward migration together with enhanced convection over subtropical western Pacific and regionally intensive precipitation including Meiyu, Changma and Baiu. The classification of fundamental large scale spatial pattern and evolutionary history of nonlinear phases of monsoon ISO provides the source of predictability for extended-range forecast of summer precipitation.

• Proceedings of the KSME Conference
• /
• 2001.06a
• /
• pp.108-113
• /
• 2001

Recently, the mechanical structures applied to many industrial products, especially in electronic products, appear to be miniaturized and complicated. This trend makes it difficult to analyze the stress distribution of those mechanical structures and generates new challenges for precise measurement of strain. Therefore, generally most of those cases largely depend on the finite element analysis. But the development of optical metrology which has the capability of non-contact, full-field and precise measurement makes it possible to solve these measuring problems. Among the optical measurement techniques, the electronic speckle pattern interferometry (ESPI) has been developed and considered as one of the most useful tools for measuring displacement and deformation. But the shortage of recognition and difficulties of measurement have limited its industrial applications in spite of its excellent capabilities. Therefore in this study, in order to enhance the industrial application of ESPI, the measurement of in-plane displacement of mechanical structure with ESPI, which is applied to washing machine and cannot be measured by strain gauges, was performed. And the verification of validity of FEA results was also done.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.15 no.8
• /
• pp.667-677
• /
• 1990

In this paper, optical inplementation of the Hopfield neural network model for two-dimensinal associative memory is described For the real-time processing of two-dimensional images, the commercial LCTVs are used as a memory mask and an input spatical light modulator. A 4-D memory matrix is realized with a 2-D mask of a matrix arrangement and the inner-products between arbitrary input pattern and memory matrix are carried out by using the multifocus hololens. The output image is then electronically thresholded and fed back to the input of the associative memory system by 2-D CCd camera. From the good experimental results for the high error correction capability, the proposed system can be applied to practical pattern recognition and machine vision systems.