• Smart Media Journal
• /
• v.9 no.2
• /
• pp.92-98
• /
• 2020

This paper proposes a vehicle license plate recognition system using light weight deep learning models without high-end server. The proposed license plate recognition system consists of 3 steps: [license plate detection]-[character area segmentation]-[character recognition]. SSD-Mobilenet was used for license plate detection, ResNet with localization was used for character area segmentation, ResNet was used for character recognition. Experiemnts using Samsung Galaxy S7 and LG Q9, accuracy showed 85.3% accuracy and around 1.1 second running time.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.444-455
• /
• 2019

The purpose of this research was to analyze technological trends of 「Military Small Giant Venture Tech-Fi Net」 from Defense Technology Information Service via social network analysis. 「Military Small Giant Venture Tech-Fi Net」, which was constituted for their fine technology for application to the military field, registered 847 technologies of 388 companies. In this research, we analyzed 847 technologies for the relations between "Military System" and "Military Technology Category" via centrality measurement, one of the social network analysis methods. The results indicate that the major technologies of domestic military small giant venture companies were "Sensor" and "ICT" for "C4I System" and "Surveillance and Reconnaissance System" and "Platform/Structure" for "Land System", "Aeronautical System" and "Naval Sea System". In contrast, we recognized inadequate technologies, such as "Propellant" and "Material" for "Missile and Ammunition system" and "Sensor" and "ICT" for "Defense System", We hope that our results and method will be conducive to the technological development of Small Giant Venture companies.

• Journal of IKEEE
• /
• v.18 no.4
• /
• pp.593-602
• /
• 2014

One of the purposes of an artificial neural netowrk(ANNet) is to implement the largest number of functions as possible with the smallest number of nodes and layers. This paper presents a CoreNet which has a multi-leveled input value and a multi-leveled output value with a 2-layered ANNet, which is the basic structure of an ANNet. I have suggested an equation for calculating the capacity of the CoreNet, which has a p-leveled input and a q-leveled output, as $a_{p,q}={\frac{1}{2}}p(p-1)q^2-{\frac{1}{2}}(p-2)(3p-1)q+(p-1)(p-2)$. I've applied this CoreNet into the simulation model 1(5)-1(6), which has 5 levels of an input and 6 levels of an output with no hidden layers. The simulation result of this model gives, the maximum 219 convergences for the number of implementable functions using the cot(${\sqrt{x}}$) input leveling method. I have also shown that, the 27 functions are implementable by the calculation of weight values(w, ${\theta}$) with the multi-threshold lines in the weight space, which are diverged in the simulation results. Therefore the 246 functions are implementable in the 1(5)-1(6) model, and this coincides with the value from the above eqution $a_{5,6}(=246)$. I also show the implementable function numbering method in the weight space.

• Journal of Biomedical Engineering Research
• /
• v.43 no.1
• /
• pp.19-26
• /
• 2022

The Comprehensive in vitro Proarrhythmic Assay(CiPA) project was launched for solving the hERG assay problem of being classified as high-risk groups even though they are low-risk drugs due to their high sensitivity. CiPA presented a protocol to predict drug toxicity using physiological data calculated based on the in-silico model. in this study, features calculated through the in-silico model are analyzed for correlation of changing action potential in the near future, and features are verified through predictive performance according to drug datasets. Using the O'Hara Rudy model modified by Dutta et al., Pearson correlation analysis was performed between 13 features(dVm/dtmax, APpeak, APresting, APD90, APD50, APDtri, Ca_peak, Ca_resting, CaD90, CaD50, CaDtri, qNet, qInward) calculated at 100 pacing, and between dVm/d_{tmax_repol} calculated at 1,000 pacing, and linear regression analysis was performed on each of the 12 training drugs, 16 verification drugs, and 28 drugs. Indicators showing high coefficient of determination(R²) in the training drug dataset were qNet 0.93, AP resting 0.83, APDtri 0.78, Ca resting 0.76, dVm/dt_max 0.63, and APD90 0.61. The indicators showing high determinants in the validated drug dataset were APDtri 0.94, APD90 0.92, APD50 0.85, CaD50 0.84, qNet 0.76, and CaD90 0.64. Indicators with high coefficients of determination for all 28 drugs are qNet 0.78, APD90 0.74, and qInward 0.59. The indicators vary in predictive performance depending on the drug dataset, and qNet showed the same high performance of 0.7 or more on the training drug dataset, the verified drug dataset, and the entire drug dataset.

• Journal of Ecology and Environment
• /
• v.38 no.4
• /
• pp.425-436
• /
• 2015

The carbon cycle came into the spotlight due to the climate change and forests are well-known for their capacity to store carbon amongst other terrestrial ecosystems. The annual organic carbon of litter production, forest floor litter layer, soil, aboveground and belowground part of plant, standing biomass, net primary production, uptake of organic carbon, soil respiration, etc. were measured in Mt. Worak in order to understand the production and carbon budget of Quercus serrata forest that are widely spread in the central and southern part of the Korean Peninsula. The total amount of organic carbon of Q. serrata forest during the study period (2010-2013) was 130.745 ton C ha^-1. The aboveground part of plant, belowground part of plant, forest floor litter layer, and organic carbon in soil was 50.041, 12.510, 4.075, and 64.119 ton C ha^-1, respectively. The total average of carbon fixation in plants from photosynthesis was 4.935 ton C ha^-1 yr^-1 and organic carbon released from soil respiration to microbial respiration was 3.972 ton C ha^-1 yr^-1. As a result, the net ecosystem production of Q. serrata forest estimated from carbon fixation and soil respiration was 0.963 ton C ha^-1 yr^-1. Therefore, it seems that Q. serrata forest can act as a sink that absorbs carbon from the atmosphere. The carbon uptake of Q. serrata forest was highest in stem of the plant and the research site had young forest which had many trees with small diameter at breast height (DBH). Consequentially, it seems that active matter production and vigorous carbon dioxide assimilation occurred in Q. serrata forest and these results have proven to be effective for Q. serrata forest to play a role as carbon storage and NEP.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.1
• /
• pp.7-13
• /
• 2022

Radio frequency fingerprinting refers to a methodology that extracts hardware-specific characteristics of a transmitter that are unintentionally embedded in a transmitted waveform. In this paper, we put forward a fingerprinting feature and deep learning structure that can identify the same type of Digital Mobile Radio(DMR) by inputting the in-phase(I) and quadrature(Q). We proposes using the magnitude in polar coordinates of I/Q as RF fingerprinting feature and a modified ResNet-1D structure that can identify them. Experimental results show that our proposed modified ResNet-1D structure can achieve recognition accuracy of 99.5% on 20 DMR.

• Journal of Ecology and Environment
• /
• v.30 no.1
• /
• pp.1-7
• /
• 2007

A model was developed to predict the effects of rising air temperature on net photosynthetic rate of Quercus mongolica stands at Mt. Paekcheok-san, Kangwon-do in South Korea. The PFD (Photon flux density) and air temperature were determined from weather data from the research site and the Daegwallyeong meteorological station and gas exchange or release responses of each tree component were measured. Using these data, we simulated the effects of increases in mean annual air temperatures above current conditions on annual $CO_2$ budget of Q. mongolica stands. If mean annual air temperature is increased by 0.5, 1.0, 1.5, 2.0, 2.5 or $3.0^{\circ}C$, annual net photosynthetic rate will be increased by 8.8, 12.8, 14.5, 12.6, 9.2 and 1.0 ton $CO_2\;ha^{-1}yr^{-1}$ respectively. Simulations indicate that changes in air temperature will have a major impact on gas exchange and release in Q. mongolica stands, resulting in a net increase in the rate of carbon fixation by standing crops.

• Asian Journal of Business Environment
• /
• v.7 no.4
• /
• pp.5-9
• /
• 2017

Purpose - Many countries rely on currency depreciation or debt-financed government spending to stimulate their economies. Currency depreciation tends to increase net exports and aggregate demand but reduce short-run aggregate supply due to higher import costs. Debt-financed government spending increases aggregate demand, but the crowding-out effect due to a higher real interest rate may reduce private spending and aggregate demand. Therefore, the net impact of currency depreciation or debt-financed government spending on equilibrium real GDP is unclear. Research design, data, and methodology - This paper examines potential impacts of real depreciation of the ringgit, more government debt as a percent of GDP and other relevant macroeconomic variables on aggregate output in Malaysia. Results - Applying the AD/AS model, this paper finds that aggregate output in Malaysia is positively associated with real appreciation during 2005.Q3-2010.Q3, real depreciation during 2010.Q4-2016.Q1, the debt-to-GDP ratio and the real stock price, negatively affected by the real lending rate and inflation expectations, and is not influenced by the real oil price. Conclusions - Real depreciation of the ringgit after 2010. Q3 or sustainable expansionary fiscal policy would be beneficial to the economy.

• Journal of IKEEE
• /
• v.15 no.4
• /
• pp.354-362
• /
• 2011

One of the unsolved problems in Artificial Neural Networks is related to the capacity of a neural network. This paper presents a CoreNet which has a multi-leveled input and a multi-leveled output as a 2-layered artificial neural network. I have suggested an equation for calculating the capacity of the CoreNet, which has a p-leveled input and a q-leveled output, as $a_{p,q}=\frac{1}{2}p(p-1)q^2-\frac{1}{2}(p-2)(3p-1)q+(p-1)(p-2)$. With an odd value of p and an even value of q, (p-1)(p-2)(q-2)/2 needs to be subtracted further from the above equation. The simulation model 1(3)-1(6) has 3 levels of an input and 6 levels of an output with no hidden layer. The simulation result of this model gives, out of 216 possible functions, 80 convergences for the number of implementable function using the cot(x) input leveling method. I have also shown that, from the simulation result, the two diverged functions become implementable by precalculating the weight values. The simulation result and the precalculation of the weight values give the same result as the above equation in the total number of implementable functions.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.27 no.3
• /
• pp.97-105
• /
• 2004

The purpose of this study is to design and implement web-based quality cost management system to measure the performance of quality improvement activities in the business firms. The designed web-based quality cost management system which is also known as WQCMS (Web-based Q-Cost Management System) utilized newly classified objective performance indicator items to measure quality related activities. Well-classified performance measurement Indicator item lists which are played key role in the system was researched and standardized according to size and business categories of enterprise. Its code standards are based from fundamental PAF Model. Developed WQCMS have ability to collect and analyze quality data generated from various different departments in the inside or outside of the enterprise without any limitations, if end-users are able to access wide area network. It provides the capability to integrate quality information from database and to generate various easy analysis reports to management's needs using built-in analysis tool modules with real-time. The proposed system was developed using Microsofi's .Net technology, ASP.NET and MS-SQL Server 2000. By web-enabling Q-cost management system, the effectiveness of the system management and utilization was realized by easiness of information Integration and economical efficiency.