• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• 제37권6호
• /
• pp.499-506
• /
• 2019

NDSM (Normalized Digital Surface Model) is key information for the detailed analysis of remote sensing data. Although NDSM can be simply obtained by subtracting a DTM (Digital Terrain Model) from a DSM (Digital Surface Model), in case of UAV (Unmanned Aerial Vehicle) data, it is difficult to get an accurate DTM due to high resolution characteristics of UAV data containing a large number of complex objects on the ground such as vegetation and urban structures. In this study, RGB-based UAV vegetation index, ExG (Excess Green) was used to extract initial seed points having low ExG values for region growing such that a DTM can be generated cost-effectively based on high resolution UAV data. For this process, local window analysis was applied to resolve the problem of erroneous seed point extraction from local low ExG points. Using the DSM values of seed points, region growing was applied to merge neighboring terrain pixels. Slope criteria were adopted for the region growing process and the seed points were determined as terrain points in case the size of segments is larger than 0.25 ㎡. Various slope criteria were tested to derive the optimized value for UAV data-based NDSM generation. Finally, the extracted terrain points were evaluated and interpolation was performed using the terrain points to generate an NDSM. The proposed method was applied to agricultural area in order to extract the above ground heights of crops and check feasibility of agricultural monitoring.

• The KIPS Transactions:PartB
• /
• 제16B권4호
• /
• pp.299-308
• /
• 2009

AAM(Active Appearance Model) is an algorithm to extract face feature points with statistical models of shape and texture information based on PCA(Principal Component Analysis). This method is widely used for face recognition, face modeling and expression recognition. However, the detection performance of AAM algorithm is sensitive to initial value and the AAM method has the problem that detection error is increased when an input image is quite different from training data. Especially, the algorithm shows high accuracy in case of closed lips but the detection error is increased in case of opened lips and deformed lips according to the facial expression of user. To solve these problems, we propose the improved AAM algorithm using lip feature points which is extracted based on a new lip detection algorithm. In this paper, we select a searching region based on the face feature points which are detected by AAM algorithm. And lip corner points are extracted by using Canny edge detection and histogram projection method in the selected searching region. Then, lip region is accurately detected by combining color and edge information of lip in the searching region which is adjusted based on the position of the detected lip corners. Based on that, the accuracy and processing speed of lip detection are improved. Experimental results showed that the RMS(Root Mean Square) error of the proposed method was reduced as much as 4.21 pixels compared to that only using AAM algorithm.

• The Korean Journal of Applied Statistics
• /
• 제32권5호
• /
• pp.693-702
• /
• 2019

Deep learning has gained popularity for the classification and prediction task. Neural network layers become deeper as more data becomes available. Saturation is the phenomenon that the gradient of an activation function gets closer to 0 and can happen when the value of weight is too big. Increased importance has been placed on the issue of saturation which limits the ability of weight to learn. To resolve this problem, Glorot and Bengio (Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, 249-256, 2010) claimed that efficient neural network training is possible when data flows variously between layers. They argued that variance over the output of each layer and variance over input of each layer are equal. They proposed a method of initialization that the variance of the output of each layer and the variance of the input should be the same. In this paper, we propose a new method of establishing initialization by adopting truncated normal distribution and truncated cauchy distribution. We decide where to truncate the distribution while adapting the initialization method by Glorot and Bengio (2010). Variances are made over output and input equal that are then accomplished by setting variances equal to the variance of truncated distribution. It manipulates the distribution so that the initial values of weights would not grow so large and with values that simultaneously get close to zero. To compare the performance of our proposed method with existing methods, we conducted experiments on MNIST and CIFAR-10 data using DNN and CNN. Our proposed method outperformed existing methods in terms of accuracy.

• Journal of the Korean Institute of Gas
• /
• 제19권1호
• /
• pp.28-37
• /
• 2015

This study presents the development of a well placement optimization model, combining an artificial neural network, which enables high-speed calculation, with a simulated annealing algorithm. The conventional FDM simulator takes excessive time when used to perform a field scale reservoir simulation. In order to solve this problem, an artificial neural network was applied to the model to allow the simulation to be executed within a short time. Also by using the given result, the optimization method, SA algorithm, was implemented to automatically select the optimal location without taking any subjective experiences into consideration. By comparing the result of the developed model with the eclipse simulator, it was found that the prediction performance of the developed model has become favorable, and the speed of calculation performance has also been improved. Especially, the optimum value was estimated by performing a sensitivity analysis for the cooling rate and the initial temperature, which is the control parameter of SA algorithm. From this result, it was verified that the calculation performance has been improved, as well. Lastly, an optimization for the well placement was performed using the model, and it concluded the optimized place for the well by selecting regions with great productivity.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• 제30권11A호
• /
• pp.1073-1080
• /
• 2005

In spite of many advantages of Meteor Burst Communications(MBC) on its transmission channel, the fact that its duty rate is less than 10 percents is a considerable deficiency of MBC. To overcome with this deficiency without paying large cost, we use a direct sequence(DS) simple reception system. This method doesn't need to add the components for acquisition or tracking the signal so that it keeps the cost efficiency of the MBC. But it has a disadvantage that its duty rate shows a large decline by noise. For this problem, we adopt the proposed system applying a positive feedback circuit. The improvement by this method is more significant when the received signal is weak. Simulation results show as tl is l00ms, then the burst length would be increased by $35\%$ when the initial value of the received signal power increases by 3 dB, t2 is 135ms. If the power increases by 6dB the burst length would be 170ms thus increased by $70\%$, which shows a great enhancement. Suppose now that tl is 50ms and the power is increased by 3dB, then t2 would be increased by $70\%$ to 85ms, and increasing the power by 6dB it would be increased by $130\%$ to 115ms.

• Journal of the Society of Naval Architects of Korea
• /
• 제33권3호
• /
• pp.35-47
• /
• 1996

A surface panel method treating a boundary-value problem of the Dirichlet type is presented to design a three dimensional body with free surface corresponding to a prescribed pressure distribution. An integral equation is derived from Green's theorem, giving a relation between total potential of known strength and the unknown local flux. Upon discretization, a system of linear simultaneous equations is formed including free surface boundary condition and is solved for an assumed geometry. The pseudo local flux, present due to the incorrect positioning of the assumed geometry, plays a role f the geometry corrector, with which the new geometry is computed for the next iteration. Sample designs for submerged spheroids and Wigley hull and carried out to demonstrate the stable convergence, the effectiveness and the robustness of the method. For the calculation of the wave resistance, normal dipoles and Rankine sources are distributed on the body surface and Rankine sources on the free surface. The free surface boundary condition is linearized with respect to the oncoming flow. Four-points upwind finite difference scheme is used to compute the free surface boundary condition. A hyperboloidal panel is adopted to represent the hull surface, which can compensate the defects of the low-order panel method. The design of a 5500TEU container carrier is performed with respect to reduction of the wave resistance. To reduce the wave resistance, calculated pressure on the hull surface is modified to have the lower fluctuation, and is applied as a Dirichlet type dynamic boundary condition on the hull surface. The designed hull form is verified to have the lower wave resistance than the initial one not only by computation but by experiment.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제16권11호
• /
• pp.7754-7760
• /
• 2015

The outer tie rod is one of the part of steering system, the optimization process was executed to find the lightweight design. The inner tie rod was considered in the optimum design of an outer tie rod. it could be closer to the test condition than in the case of considering outer tie rod only. The aluminum forging material was considered as a weight reduction proposal. The target of optimization was the shape of the minimum weight to resist at the load of buckling. RSM and Kriging interpolation method were applied as a optimization method to consider the nonlinear shape optimization problem. Then, 16.3%, 16.6% of weight reduction was obtained from the result comparing with that of the initial model. The results of meta model optimization was compared with that of finite element method. The error values of buckling load estimation were 2.6%, 2.04%. and those of weight estimation were 0.17%, 0.13%. Therefore, it seemed that the result of Kriging model could be obtained closer to optimum value than that of RSM model.

• Korean Journal of Heritage: History & Science
• /
• 제49권4호
• /
• pp.114-125
• /
• 2016

Lacquer is one of the most widely used natural resin in East Asia since Neolithic Age. As a major ingredient of lacquerware, lacquer is waterproof, insect-proof and rot-proof to be durable and anti-abrasion, generally utilized for mainly painting purpose. According to lacquerware excavated from several sites of Japan and China, lacquerware seems to appear in Neolithic Age. On the other hand in case of excavation research in Korea, lacquerwares are found after the Bronze Age. The initial purpose of lacquer is estimated to be adhesive, regarding the literatures mentioning bitumen(Yeoksceong ), animal glue(Kyeo) and egg alumen(Nanbaek). Especially piece of jar coffin unearthed in Pyeongtaek Daechu-ri site had trace of restoration by lacquer and hemp as an evidence of lacquer for adhesion in Pre-Three Kingdoms period. Since then a trend to restore the broken ceramics with lacquer and decorate with golden foil lasted especially in Joseon Dynasty. In the field of gold plated lacquer method, mother-of-pearl inlaying technique for lacquerware and restoration of buildings, lacquer is still used as adhesive. Due to matter of reversibility lacquer is being avoided for conservation and restoration of cultural heritage. Lacquer as a traditional material for adhesive since ancient times, however, has advantage in adhesion strength and durability. Because synthetic resin adhesive has problem of emission of volatile organic compounds and aging over time, lacquer receives attention recently. On the contrary, by combination adhesive from mixing lacquer and animal glue, already proved the possibility of applicability and chemical modification. A research to utilize lacquer as modern paint or functional material is also conducted continuously also in China and Japan. To put traditional material into practical use and modernize, chemical research from the molecular level of the lacquer is necessary in the near future.

• Smart Media Journal
• /
• 제9권4호
• /
• pp.102-108
• /
• 2020

In this paper, we propose a real-time method of detecting hand motions and extracting the signal frame induced by EF(Electric Field) sensors. The signal induced by hand motion includes not only noises caused by various environmental sources as well as sensor's physical placement, but also different initial off-set conditions. Thus, it has been considered as a challenging problem to detect the motion signal and extract the motion frame automatically in real-time. In this study, we remove the PLN(Power Line Noise) using LPF with 10Hz cut-off and successively apply MA(Moving Average) filter to obtain clean and smooth input motion signals. To sense a hand motion, we use two thresholds(positive and negative thresholds) with offset value to detect a starting as well as an ending moment of the motion. Using this approach, we can achieve the correct motion detection rate over 98%. Once the final motion frame is determined, the motion signals are normalized to be used in next process of classification or recognition stage such as LSTN deep neural networks. Our experiment and analysis show that our proposed methods produce better than 98% performance in correct motion detection rate as well as in frame-matching rate.

• Magazine of the Korean Society of Agricultural Engineers
• /
• 제26권1호
• /
• pp.52-72
• /
• 1984

This study was performed to obtain the basic data which can be applied to the use of epoxy resin mortars. The data was based on the properties of epoxy resin mortars depending upon various mixing ratios to compare those of cement mortar. The resin which was used at this experiment was Epi-Bis type epoxy resin which is extensively being used as concrete structures. In the case of epoxy resin mortar, mixing ratios of resin to fine aggregate were 1: 2, 1: 4, 1: 6, 1: 8, 1:10, 1 :12 and 1:14, but the ratio of cement to fine aggregate in cement mortar was 1 : 2.5. The results obtained are summarized as follows; 1.When the mixing ratio was 1: 6, the highest density was 2.01 g/cm$^3$, being lower than 2.13 g/cm$^3$ of that of cement mortar. 2.According to the water absorption and water permeability test, the watertightness was shown very high at the mixing ratios of 1: 2, 1: 4 and 1: 6. But then the mixing ratio was less than 1 : 6, the watertightness considerably decreased. By this result, it was regarded that optimum mixing ratio of epoxy resin mortar for watertight structures should be richer mixing ratio than 1: 6. 3.The hardening shrinkage was large as the mixing ratio became leaner, but the values were remarkably small as compared with cement mortar. And the influence of dryness and moisture was exerted little at richer mixing ratio than 1: 6, but its effect was obvious at the lean mixing ratio, 1: 8, 1:10,1:12 and 1:14. It was confirmed that the optimum mixing ratio for concrete structures which would be influenced by the repeated dryness and moisture should be rich mixing ratio higher than 1: 6. 4.The compressive, bending and splitting tensile strenghs were observed very high, even the value at the mixing ratio of 1:14 was higher than that of cement mortar. It showed that epoxy resin mortar especially was to have high strength in bending and splitting tensile strength. Also, the initial strength within 24 hours gave rise to high value. Thus it was clear that epoxy resin was rapid hardening material. The multiple regression equations of strength were computed depending on a function of mixing ratios and curing times. 5.The elastic moduli derived from the compressive stress-strain curve were slightly smaller than the value of cement mortar, and the toughness of epoxy resin mortar was larger than that of cement mortar. 6.The impact resistance was strong compared with cement mortar at all mixing ratios. Especially, bending impact strength by the square pillar specimens was higher than the impact resistance of flat specimens or cylinderic specimens. 7.The Brinell hardness was relatively larger than that of cement mortar, but it gradually decreased with the decline of mixing ratio, and Brinell hardness at mixing ratio of 1 :14 was much the same as cement mortar. 8.The abrasion rate of epoxy resin mortar at all mixing ratio, when Losangeles abation testing machine revolved 500 times, was very low. Even mixing ratio of 1 :14 was no more than 31.41%, which was less than critical abrasion rate 40% of coarse aggregate for cement concrete. Consequently, the abrasion rate of epoxy resin mortar was superior to cement mortar, and the relation between abrasion rate and Brinell hardness was highly significant as exponential curve. 9.The highest bond strength of epoxy resin mortar was 12.9 kg/cm$^2$ at the mixing ratio of 1:2. The failure of bonded flat steel specimens occurred on the part of epoxy resin mortar at the mixing ratio of 1: 2 and 1: 4, and that of bonded cement concrete specimens was fond on the part of combained concrete at the mixing ratio of 1 : 2 ,1: 4 and 1: 6. It was confirmed that the optimum mixing ratio for bonding of steel plate, and of cement concrete should be rich mixing ratio above 1 : 4 and 1 : 6 respectively. 10.The variations of color tone by heating began to take place at about 60˚C, and the ultimate change occurred at 120˚C. The compressive, bending and splitting tensile strengths increased with rising temperature up to 80˚ C, but these rapidly decreased when temperature was above 800 C. Accordingly, it was evident that the resistance temperature of epoxy resin mortar was about 80˚C which was generally considered lower than that of the other concrete materials. But it is likely that there is no problem in epoxy resin mortar when used for unnecessary materials of high temperature resistance. The multiple regression equations of strength were computed depending on a function of mixing ratios and heating temperatures. 11.The susceptibility to chemical attack of cement mortar was easily affected by inorganic and organic acid. and that of epoxy resin mortar with mixing ratio of 1: 4 was of great resistance. On the other hand, when mixing ratio was lower than 1 : 8 epoxy resin mortar had very poor resistance, especially being poor resistant to organicacid. Therefore, for the structures requiring chemical resistance optimum mixing of epoxy resin mortar should be rich mixing ratio higher than 1: 4.