• Journal of the Korea Institute of Military Science and Technology
• /
• v.22 no.2
• /
• pp.278-286
• /
• 2019

Conventional methods for selecting jamming techniques in electronic warfare are based on libraries in which a list of jamming techniques for radar signals is recorded. However, the choice of jamming techniques by the library is limited when modified signals are received. In this paper, we propose a method to predict the jamming technique for radar signals by using deep learning methods. Long short-term memory(LSTM) is a deep running method which is effective for learning the time dependent relationship in sequential data. In order to determine the optimal LSTM model structure for jamming technique prediction, we test the learning parameter values that should be selected, such as the number of LSTM layers, the number of fully-connected layers, optimization methods, the size of the mini batch, and dropout ratio. Experimental results demonstrate the competent performance of the LSTM model in predicting the jamming technique for radar signals.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.9
• /
• pp.2342-2352
• /
• 1995

Deposition of flame-synthesized silica particles onto a target is utilized in optical fiber preform fabrication processes. The particles are convected and deposited onto the target. Falkner-Skan wedge flow was chosen as the particle laden flow. Typically the particles are polydisperse in size and follow a lognormal size distribution. Brownian diffusion, thermophoresis, and coagulation of the particles were considered and effects of these phenomena on particle deposition were studied. A moment model was developed in order to predict the particle number density and the particle size distribution simultaneously. Particle deposition with various wedge configurations was examined for conditions selected for a typical VAD process. When coagulation was considered, mean particle size and its standard deviation increased and particle number density decreased, compared to the case without coagulation. These results proved the fact that coagulation effect expands particle size distribution. The results were discussed with characteristics of thermal and diffusion boundary layers. As the boundary layers grow in thickness, overall temperature and concentration gradients decrease, resulting in decrease of deposition rate and increase of particle residence time in the flow and thus coagulation effect.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.37 no.3
• /
• pp.111-122
• /
• 2021

Purpose: The purpose of this in vitro study was to investigate the flexural strength and translucency of three layers in 5Y-ZP and to assess the effect of additional firings on these properties. Materials and Methods: Sintered zirconia blocks were sectioned according to three layers : incisal, transition, and body. Disc-shaped specimens were fabricated from each layer. The diameter of specimens was 15.0 mm and each thickness of specimens for biaxial flexural strength test and translucency was 1.2 mm and 1.0 mm. The specimens were classified into subgroups according to the number of firing (0, 1, and 3 times; n = 10/subgroup) and the additional firings were performed under 900℃ using a furnace. Biaxial flexural strength and translucency was measured using universal testing machine and uv-vis spectrophotometer. X-ray diffraction (XRD) analysis was used for measurement of the phase identification. One-way ANOVA, Tukey HSD test were performed (α = 0.05). Results: There was no significant difference in flexural strength between the three layers (P > 0.05), while there was significant difference in translucency between different layers (P < 0.05). The flexural strength of incisal and transition layer was decreased by the single additional firing, and the three additional firings significantly decreased the flexural strength of three layers. The translucency of layer was decreased by additional firings except the body layer. The XRD patterns of all groups were similar. Conclusion: Three layers of 5Y-ZP were different only in translucency. Additional firings affected the flexural strength and translucency differently depending on the layers but crystalline phases were not changed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.7
• /
• pp.1457-1464
• /
• 2011

In this paper, we go one step further in that the number of competitive layers is not determined by experience but can be determined by fuzzy control rules based on input pattern information. In our method, we design a set of membership functions and corresponding rules and used Max-Min reasoning proposed by Mamdani. Also, we use centroid method as a defuzzification. In experiment that has various patterns of English inputs, this new method works beautifully to determine the number of competitive layers and also efficient in overall accuracy as a result.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.1
• /
• pp.92-98
• /
• 2019

In this paper, we propose a deep learning system based on morphological neural network(MNN). The deep learning layers are morphological operation layer, pooling layer, ReLU layer, and the fully connected layer. The operations used in morphological layer are erosion, dilation, and edge detection, etc. Unlike CNN, the number of hidden layers and kernels applied to each layer is limited in MNN. Because of the reduction of processing time and utility of VLSI chip design, it is possible to apply MNN to various mobile embedded systems. MNN performs the edge and shape detection operations with a limited number of kernels. Through experiments using database images, it is confirmed that MNN can be used as a deep learning system and its performance.

• Journal of Climate Change Research
• /
• v.7 no.1
• /
• pp.49-57
• /
• 2016

In the present study, the time series of the number of days on which temperatures were not higher than $-5^{\circ}C$ in winter in Seoul was analyzed. The results showed a decreasing tendency until recently. Statistical change-point analysis was conducted to examine whether climate regime shifts existed in this time series. According to the results, the number of days on which temperatures were not higher than $-5^{\circ}C$ in winter in Seoul drastically decreased since 1988. Therefore, to find out the reason for the recent decrease in the number of days, differences between the means of large-scale environments in winder during 1988~2010 and those during 1974~1987 were analyzed. In all layers of the troposphere, anomalous anticyclones developed in regions around the Korean Peninsula and thus the Korean Peninsula was affected by westerlies or south-westerlies. This was associated with the recent a little further northward development of western North Pacific subtropical high. Therefore, environments good for warm and humid air to flow into the Korean Peninsula were formed. To examine whether relatively warm and humid air actually flowed into the Korean Peninsula recently, temperatures and specific humidity in all layers in the troposphere were analyzed and according to the results the Korean Peninsula showed warm and humid anomalies. In the analyses of sensible heat net flux and maximum temperatures at a height of 2 m that can be felt by humans, the East Asia Continent including the Korean Peninsula showed positive anomalies.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.1
• /
• pp.73-80
• /
• 2016

The Car License Plate Recognition(: CLPR) is required in searching the hit-and-run car, measuring the traffic density, investigating the traffic accidents as well as in pursuing vehicle crimes according to the increasing in number of vehicles. The captured images on the real environment of the CLPR is contaminated not only by snow and rain, illumination changes, but also by the geometrical distortion due to the pose changes between camera and car at the moment of image capturing. We propose homographic transformation and intensity histogram of vertical image projection so as to transform the distorted input to the original image and cluster the character and number, respectively. Especially, in this paper, the Multilayer Perceptron Algorithm(: MLP) in the CLPR is used to not only recognize the charcters and car license plate, but also determine the optimized ratio among the number of input, hidden and output layers by the real experimental result.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.32 no.6
• /
• pp.516-523
• /
• 2020

Since the aging of coastal structures have been increased, the researches about the reinforcements of the existing aged structures are needed. Especially, the existing armor units placed on rubble mound structures should satisfy the stability against the increased design wave conditions. However the researches about these design problems have not been performed. In this study, the hydraulic model tests to investigate the stability number about the additionally placed armor unit were conducted. The main armor unit is a Tetrapod. The test results showed that the stability number (Kd) for additionally placed armor units(Tetrapod) increased up to maximum 10% comparing with that for 2 layers tetrapod (Kd = 8) within these test conditions with the pattern placing for existing armor layers and the stable armor layer slope for the non overtopping condition.

• Archives of Metallurgy and Materials
• /
• v.64 no.2
• /
• pp.491-494
• /
• 2019

In this study, we investigate the mechanical behavior of each skin layer, in terms of the nominal stress-strain curve by uniaxial tensile tests using specimens of porcine skin in two forms: dermis containing epidermis, and all three layers. All tests were performed under cyclic loading at the constant strain rate of 10^-3 s^-1 at ambient temperature. To measure the precise initial cross-sectional areas of each layer, the thickness of each skin layer was quantified by counting the number of pixels on the photo-image using image-processing software. In the tensile test, force-strain curves of the total skin and dermis with epidermis were obtained. Subsequently, a rule of mixtures was applied to determine the nonlinear mechanical properties of the hypodermis layer. In conclusion, we could define the uniaxial tensile behavior of the hypodermis, and additionally predict the weight effect of the dermis and hypodermis layers in the tensile test.

• Advances in concrete construction
• /
• v.10 no.1
• /
• pp.35-48
• /
• 2020

This paper deals with a comparative study among three different rehabilitation techniques, namely, (i) carbon fibre reinforced polymer (CFRP), (ii) glass fibre reinforced polymer (GFRP) and (iii) ferrocement on the flexural strengthening of reinforced cement concrete (RCC) beams. As these different techniques have to be compared on a level playing field, tensile coupon tests have been carried out initially for GFRP, CFRP and ferrocement and the number of layers required in each of these composites in terms of the tensile strength. It was found that for the selected constituents of the composites, one layer of CFRP was equivalent to three layers of GFRP and five layers of wiremesh reinforcement in ferrocement. Rehabilitation of RCC beams using these equivalent laminates shows that all the three composites performed in a similar way and are comparable. The parameters selected in this study were (i) the strengthening material and (ii) the level of pre-distress induced to the beams prior to the rehabilitation. It was noticed that, as the levels of pre-distress decreases, the percentage attainment of flexural capacity and flexural stiffness of the rehabilitated beams increases for all the three selected composites used for rehabilitation. Load-deflection behavior, failure modes, energy absorption capacity, displacement ductility and curvature ductility were compared among these composites and at different distress levels for each composite. The results indicate that ferrocement showed a better performance in terms of ductility than other FRPs, and between the FRPs, GFRP exhibited a better ductility than the CFRP counterpart.