• Applied Chemistry for Engineering
• /
• v.22 no.2
• /
• pp.230-234
• /
• 2011

In this work, the primary material for a single layered hardmask which can afford a spin-on process was prepared by the minture of organic and inorganic sources. The preparation of hybrid polymer was attempted by esterification from silanol terminated siloxane compounds and acetonide-2,2-bis(methoxy)propionic acid. The optical, thermal and morphological properties of the test hardmask film was examined in terms of cross-linking agent and additives. In addition, the etch rate of hardmask film and photo resist layer were compared. The hybrid polymer prepared from organic and inorganic materials was found to be useful for hardmask film to form the nano-patterns.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.2
• /
• pp.109-118
• /
• 2022

GPR (Ground Penetrating Radar), developed as a technology for geotechnical investigations such as sinkhole exploration, was used limitedly as a method to resolve undetectable lines in underground facility exploration. To improve the accuracy of underground facility data, the government made it possible to explore underground facilities using a non-metallic pipeline probe from July 2022. However, GPR has a problem in that the exploration rate is lowered in the soil with high moisture content, such as soft soil, such as clay layer, and there is a lot of variation in long-term accuracy. In this study, as a way to improve the accuracy of exploration considering the characteristics of GPR and the environment of underground facilities, we propose a GPR exploration method for underground facilities using permittivity constant correction and pattern analysis of GPR image data. Through this study, the accuracy of underground facility exploration and high reproducibility were derived as a result of field verification applying GPR frequency band and heterogeneous GPR.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1649-1654
• /
• 2021

In this paper, we introduce a regularization of long short-term memory (LSTM) based fall detection system using TensorFlow that can detect falls that can occur in the elderly. Fall detection uses data from a 3-axis acceleration sensor attached to the body of an elderly person and learns about a total of 7 behavior patterns, each of which is a pattern that occurs in daily life, and the remaining 3 are patterns for falls. During training, a normalization process is performed to effectively reduce the loss function, and the normalization performs a maximum-minimum normalization for data and a L2 regularization for the loss function. The optimal regularization conditions of LSTM using several falling parameters obtained from the 3-axis accelerometer is explained. When normalization and regularization rate λ for sum vector magnitude (SVM) are 127 and 0.00015, respectively, the best sensitivity, specificity, and accuracy are 98.4, 94.8, and 96.9%, respectively.

• Journal of the Microelectronics and Packaging Society
• /
• v.24 no.4
• /
• pp.39-46
• /
• 2017

This paper presents the fabrication of ceramic insulation layer on metallic heat spreading substrate, i.e. an insulated metal substrate, for planar type heater. Aluminum alloy substrate is preferred as a heat spreading panel due to its high thermal conductivity, machinability and the light weight for the planar type heater which is used at the thermal treatment process of semiconductor device and display component manufacturing. An insulating layer made of ceramic dielectric film that is stable at high temperature has to be coated on the metallic substrate to form a heating element circuit. Two technical issues are raised at the forming of ceramic insulation layer on the metallic substrate; one is delamination and crack between metal and ceramic interface due to their large differences in thermal expansion coefficient, and the other is electrical breakdown due to intrinsic weakness in dielectric or structural defects. In this work, to overcome those problem, selected metal oxide buffer layers were introduced between metal and ceramic layer for mechanical matching, enhancing the adhesion strength, and multi-coating method was applied to improve the film quality and the dielectric breakdown property.

• Journal of Sensor Science and Technology
• /
• v.9 no.3
• /
• pp.224-232
• /
• 2000

Aluminum metal process in surface micromachining enables to fabricate Al electrodes or Al structures, which improve electrical characteristics by reducing contact- and line-resistance or makes the whole process to be simple by using oxide as sacrificial layer. However, it is not possible to use conventional sacrificial layer etching process, because HF solution attacks aluminum as well as sacrificial oxide. The mixed solution of BHF and glycerine as an alternative shows the adequate properties to meet with this end. The exact etching properties, however, are sensitively depends on the geometry of the released structure, because the most etching process of sacrificial layer proceeds to the lateral direction in narrow space. Also, the surface roughness of aluminum affects to the etching characteristics. This paper reports experimental results on the effect of microstructure and surface roughness of aluminum to the etching properties. Considering these effects, we propose the optimized etching condition, which can be used practically for the fabrication of aluminum electrodes and microstructures by using standard surface micromachining process without modification or additional process.

• 한국해양학회지
• /
• v.22 no.1
• /
• pp.43-56
• /
• 1987

A study on sedimentation, geochemical behavior and seismic stratigrapht of the continental shelf sediments along the Korea Strait and a part of south and southeast offshore area of the Korea Peninsula was carried out. In the inner shelf floor with depth ranging up to 80m zonal distribution patterns of mud, sandy silt, and silty sand were observed. In the outer shelf, however, coarse sandy sediments are dominant, and shills and gravels were frequently observed. These observations seem to confirm the Holocene sedimentary processes on the continental shelves off the south, south to east coasts of Korea discussed by Park (1985 and 1986) and Park and Choi (1986). The suface sediments (upper most 5cm thick)from selected 9 stations were analyzed for Al,Mn, Fe,Cr,Ni,Cu,Zn and Pb in order to study geochemical behavior of the sediments in the study area. All data were normalized to Al to com,pensate the size effect of the sediments.In general,inner shelf sediments show slight enrichment compared to the outer shelf sediments.In particular,Pb and Zn show heavy enrichment in most of the sediments.to degrees comparable to those observed at the polluted Kwangyang and Masan Bay sediments.Thus,it is considered that rapid migration or movement of fine-grained sediments in the study area does exist. Three seismic stratigraphic units were analyzed based on the seismic records.The acoustic basement the lower sedimentary deposit(B)and the upper deposit(A)were observed.The strong reflectivity R,in particular, between unit A and B is considered to be an erosinal unconformity during the last Glacial time.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.3
• /
• pp.285-290
• /
• 2019

This study aims to compare the energy consumption characteristics of high-rise and general apartments and propose policy implications in Incheon City where high-rise apartments are planned. The method of analysis is to select the cases, to conduct field survey, drawing review, analysis of Electric Architectural administration Information System. The study derived the current status of energy consumption in high-rise and general apartment complexes located in the same region, Yeonsu-gu Incheon City, and performed comparative analysis on their characteristics. First, electrical energy in the high-rise apartment complexes was consumed excessively, by 1.63 to 2.5 times more than that of the general apartment complexes. Second, the gas energy usage in the high-rise apartment complexes was higher than that of the general complexes, by 1.09 to 1.2 times. Third, the energy consumption per unit area in the high-rises was also higher, by 1.042 to 1.3 times. As individual elements such as incomes, living standards, and life patterns of the residents affect energy consumption, the high-rise apartment complex consumed more energy per unit area than the general apartment complex did. However, this study did not consider the elements of energy expenditure and satisfaction level, which are the limitations of this research.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.10
• /
• pp.2028-2042
• /
• 1994

Properly defining signal and noise in a self-organizing system like ART(Adaptive Resonance Theory) neural network model raises a number of subtle issues. Pattern context must enter the definition so that input features, treated as irrelevant noise when they are embedded in a given input pattern, may be treated as informative signals when they are embedded in a different input pattern. The ATR automatically self-scales their computational units to embody context and learning dependent definitions of a signal and noise and there is no problem in categorizing input pattern that have features similar in nature. However, when we have imput patterns that have features that are different in size and nature, the use of only one vigilance parameter is not enough to differentiate a signal from noise for a good categorization. For example, if the value fo vigilance parameter is large, then noise may be processed as an informative signal and unnecessary categories are generated: and if the value of vigilance parameter is small, an informative signal may be ignored and treated as noise. Hence it is no easy to achieve a good pattern categorization. To overcome such problems, a Coupled-ART neural network capable of modularized categorization of patterns is proposed. The Coupled-ART has two layer of tightly coupled modules. the upper and the lower. The lower layer processes the global features of a pattern and the structural features, separately in parallel. The upper layer combines the categorized outputs from the lower layer and categorizes the combined output, Hence, due to the modularized categorization of patterns, the Coupled-ART classifies patterns more efficiently than the ART1 model.

• Journal of The Korean Association of Information Education
• /
• v.1 no.1
• /
• pp.28-37
• /
• 1997

This paper presents a new neural network based on fuzzy set and its application to invariant character recognition. The fuzzy neural network consists of five layers. The results of simulation show that the network can recognize characters in the case of distortion, translation, rotation and different sizes of handwritten characters and even with noise(8${\sim}$30%)). Translation, distortion, different sizes and noise are achieved by layer L2 and rotation invariant by layer L5. The network can recognize 108 examples of training with 100% recognition rate when they are shifted in eight directions by 1 pixel and 2 pixels. Also, the network can recognize all the distorted characters with 100% recognition rate. The simulations show that the test patterns cover a ${\pm}20^{\circ}$ range of rotation correctly. The proposed network can also recall correctly all the learned characters with 100% recognition rate. The proposed network is simple and its learning and recall speeds are very fast. This network also works for the segmentation and recognition of handwritten characters.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.4
• /
• pp.60-68
• /
• 1994

In this paper we propose a new Hybrid learning rule applied to multilayer feedforward neural networks, which is constructed by combining Hebbian learning rule that is a good feature extractor and Back-Propagation(BP) learning rule that is an excellent classifier. Unlike the BP rule used in multi-layer perceptron(MLP), the proposed Hybrid learning rule is used for uptate of all connection weights except for output connection weigths becase the Hebbian learning in output layer does not guarantee learning convergence. To evaluate the performance, the proposed hybrid rule is applied to classifier problems in two dimensional space and shows better performance than the one applied only by the BP rule. In terms of learning speed the proposed rule converges faster than the conventional BP. For example, the learning of the proposed Hybrid can be done in 2/10 of the iterations that are required for BP, while the recognition rate of the proposed Hybrid is improved by about $0.778\%$ at the peak.