• Korean Journal of Remote Sensing
• /
• v.39 no.2
• /
• pp.183-192
• /
• 2023

The remotely sensed data, such as satellite imagery and aerial photos, can be used to extract and detect some objects in the image through image interpretation and processing techniques. Significantly, the possibility for utilizing digital map updating and land monitoring has been increased through automatic object detection since spatial resolution of remotely sensed data has improved and technologies about deep learning have been developed. In this paper, we tried to extract plastic greenhouses into aerial orthophotos by using fully convolutional densely connected convolutional network (FC-DenseNet), one of the representative deep learning models for semantic segmentation. Then, a quantitative analysis of extraction results had performed. Using the farm map of the Ministry of Agriculture, Food and Rural Affairsin Korea, training data was generated by labeling plastic greenhouses into Damyang and Miryang areas. And then, FC-DenseNet was trained through a training dataset. To apply the deep learning model in the remotely sensed imagery, instance norm, which can maintain the spectral characteristics of bands, was used as normalization. In addition, optimal weights for each band were determined by adding attention modules in the deep learning model. In the experiments, it was found that a deep learning model can extract plastic greenhouses. These results can be applied to digital map updating of Farm-map and landcover maps.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.1
• /
• pp.105-122
• /
• 2022

According to the statistics about the fatal crashes that have occurred on the expressways for the last 5 years, those who died on the shoulders of the road has been as 3 times high as the others who died on the expressways. It suggests that the crashes on the shoulders of the road should be fatal, and that it would be important to prevent the traffic crashes by cracking down on the vehicles intruding the shoulders of the road. Therefore, this study proposed a method to detect a vehicle that violates the shoulder lane by using the Faster R-CNN. The vehicle was detected based on the Faster R-CNN, and an additional reading module was configured to determine whether there was a shoulder violation. For experiments and evaluations, GTAV, a simulation game that can reproduce situations similar to the real world, was used. 1,800 images of training data and 800 evaluation data were processed and generated, and the performance according to the change of the threshold value was measured in ZFNet and VGG16. As a result, the detection rate of ZFNet was 99.2% based on Threshold 0.8 and VGG16 93.9% based on Threshold 0.7, and the average detection speed for each model was 0.0468 seconds for ZFNet and 0.16 seconds for VGG16, so the detection rate of ZFNet was about 7% higher. The speed was also confirmed to be about 3.4 times faster. These results show that even in a relatively uncomplicated network, it is possible to detect a vehicle that violates the shoulder lane at a high speed without pre-processing the input image. It suggests that this algorithm can be used to detect violations of designated lanes if sufficient training datasets based on actual video data are obtained.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.1-8
• /
• 2023

If there is a defect in the wheel bearing, which is a major part of the car, it can cause problems such as traffic accidents. In order to solve this problem, big data is collected and monitoring is conducted to provide early information on the presence or absence of wheel bearing failure and type of failure through predictive diagnosis and management technology. System development is needed. In this paper, to implement such an intelligent wheel hub bearing maintenance system, we develop an embedded system equipped with sensors for monitoring reliability and soundness and algorithms for predictive diagnosis. The algorithm used acquires vibration signals from acceleration sensors installed in wheel bearings and can predict and diagnose failures through big data technology through signal processing techniques, fault frequency analysis, and health characteristic parameter definition. The implemented algorithm applies a stable signal extraction algorithm that can minimize vibration frequency components and maximize vibration components occurring in wheel bearings. In noise removal using a filter, an artificial intelligence-based soundness extraction algorithm is applied, and FFT is applied. The fault frequency was analyzed and the fault was diagnosed by extracting fault characteristic factors. The performance target of this system was over 12,800 ODR, and the target was met through test results.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.1-25
• /
• 2020

In this paper, we suggest an application system architecture which provides accurate, fast and efficient automatic gasometer reading function. The system captures gasometer image using mobile device camera, transmits the image to a cloud server on top of private LTE network, and analyzes the image to extract character information of device ID and gas usage amount by selective optical character recognition based on deep learning technology. In general, there are many types of character in an image and optical character recognition technology extracts all character information in an image. But some applications need to ignore non-of-interest types of character and only have to focus on some specific types of characters. For an example of the application, automatic gasometer reading system only need to extract device ID and gas usage amount character information from gasometer images to send bill to users. Non-of-interest character strings, such as device type, manufacturer, manufacturing date, specification and etc., are not valuable information to the application. Thus, the application have to analyze point of interest region and specific types of characters to extract valuable information only. We adopted CNN (Convolutional Neural Network) based object detection and CRNN (Convolutional Recurrent Neural Network) technology for selective optical character recognition which only analyze point of interest region for selective character information extraction. We build up 3 neural networks for the application system. The first is a convolutional neural network which detects point of interest region of gas usage amount and device ID information character strings, the second is another convolutional neural network which transforms spatial information of point of interest region to spatial sequential feature vectors, and the third is bi-directional long short term memory network which converts spatial sequential information to character strings using time-series analysis mapping from feature vectors to character strings. In this research, point of interest character strings are device ID and gas usage amount. Device ID consists of 12 arabic character strings and gas usage amount consists of 4 ~ 5 arabic character strings. All system components are implemented in Amazon Web Service Cloud with Intel Zeon E5-2686 v4 CPU and NVidia TESLA V100 GPU. The system architecture adopts master-lave processing structure for efficient and fast parallel processing coping with about 700,000 requests per day. Mobile device captures gasometer image and transmits to master process in AWS cloud. Master process runs on Intel Zeon CPU and pushes reading request from mobile device to an input queue with FIFO (First In First Out) structure. Slave process consists of 3 types of deep neural networks which conduct character recognition process and runs on NVidia GPU module. Slave process is always polling the input queue to get recognition request. If there are some requests from master process in the input queue, slave process converts the image in the input queue to device ID character string, gas usage amount character string and position information of the strings, returns the information to output queue, and switch to idle mode to poll the input queue. Master process gets final information form the output queue and delivers the information to the mobile device. We used total 27,120 gasometer images for training, validation and testing of 3 types of deep neural network. 22,985 images were used for training and validation, 4,135 images were used for testing. We randomly splitted 22,985 images with 8:2 ratio for training and validation respectively for each training epoch. 4,135 test image were categorized into 5 types (Normal, noise, reflex, scale and slant). Normal data is clean image data, noise means image with noise signal, relfex means image with light reflection in gasometer region, scale means images with small object size due to long-distance capturing and slant means images which is not horizontally flat. Final character string recognition accuracies for device ID and gas usage amount of normal data are 0.960 and 0.864 respectively.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.08a
• /
• pp.80-81
• /
• 2012

Recently, one of the critical issues in the etching processes of the nanoscale devices is to achieve ultra-high aspect ratio contact (UHARC) profile without anomalous behaviors such as sidewall bowing, and twisting profile. To achieve this goal, the fluorocarbon plasmas with major advantage of the sidewall passivation have been used commonly with numerous additives to obtain the ideal etch profiles. However, they still suffer from formidable challenges such as tight limits of sidewall bowing and controlling the randomly distorted features in nanoscale etching profile. Furthermore, the absence of the available plasma simulation tools has made it difficult to develop revolutionary technologies to overcome these process limitations, including novel plasma chemistries, and plasma sources. As an effort to address these issues, we performed a fluorocarbon surface kinetic modeling based on the experimental plasma diagnostic data for silicon dioxide etching process under inductively coupled C4F6/Ar/O2 plasmas. For this work, the SiO2 etch rates were investigated with bulk plasma diagnostics tools such as Langmuir probe, cutoff probe and Quadruple Mass Spectrometer (QMS). The surface chemistries of the etched samples were measured by X-ray Photoelectron Spectrometer. To measure plasma parameters, the self-cleaned RF Langmuir probe was used for polymer deposition environment on the probe tip and double-checked by the cutoff probe which was known to be a precise plasma diagnostic tool for the electron density measurement. In addition, neutral and ion fluxes from bulk plasma were monitored with appearance methods using QMS signal. Based on these experimental data, we proposed a phenomenological, and realistic two-layer surface reaction model of SiO2 etch process under the overlying polymer passivation layer, considering material balance of deposition and etching through steady-state fluorocarbon layer. The predicted surface reaction modeling results showed good agreement with the experimental data. With the above studies of plasma surface reaction, we have developed a 3D topography simulator using the multi-layer level set algorithm and new memory saving technique, which is suitable in 3D UHARC etch simulation. Ballistic transports of neutral and ion species inside feature profile was considered by deterministic and Monte Carlo methods, respectively. In case of ultra-high aspect ratio contact hole etching, it is already well-known that the huge computational burden is required for realistic consideration of these ballistic transports. To address this issue, the related computational codes were efficiently parallelized for GPU (Graphic Processing Unit) computing, so that the total computation time could be improved more than few hundred times compared to the serial version. Finally, the 3D topography simulator was integrated with ballistic transport module and etch reaction model. Realistic etch-profile simulations with consideration of the sidewall polymer passivation layer were demonstrated.

• Journal of Korean Society of Environmental Engineers
• /
• v.35 no.12
• /
• pp.937-942
• /
• 2013

Fluid generated within the sonic or ultrasonic waves are reflected by the wall, while the opposite direction forming a predetermined sound wave to the acoustic standing wave is referred to. In this study, the frequency of 1.0 MHz and 2.0 MHz acoustic standing wave generation module is installed in a continuous particle separation device, the laminar flow of influent, taking into account the hydraulic retention time (HRT) in accordance with changes in particle separation characteristics investigated. Operation of a standing wave in the particle separation device about $1.3{\sim}2.8^{\circ}C$ temperature is increased, but did not significantly affect the formation of standing waves. During operation, the HRT 1 hr frequency 1.0 MHz 2 hr, 4 hr longer as the particle separation efficiency (turbidity) were 64.1%, 70.0%, 74.3% and, 2.0 MHz has 58.0%, respectively, depending on HRT, 61.8%, 70.7% in the respectively. That is, the same frequency, the HRT treatment efficiency is 10% or more, depending on differences in generation and, 1.0 MHz frequency, 2 hr, 2.0 MHz 4 hr at about 70% or more of the processing efficiency can be maintained. Frequency of 1.0 MHz and 2.0 MHz operation at the same time, as a result, HRT 1 hr, 2 hr, 4 hr particle separation efficiency of 63.8%, respectively, 70.6%, 77.6%, rather than the generation of standing waves appear continuous HRT is affecting a lot of particles to separate could know.

• Journal of Intelligence and Information Systems
• /
• v.20 no.2
• /
• pp.59-71
• /
• 2014

System trading is becoming more popular among Korean traders recently. System traders use automatic order systems based on the system generated buy and sell signals. These signals are generated from the predetermined entry and exit rules that were coded by system traders. Most researches on system trading have focused on designing profitable entry and exit rules using technical indicators. However, market conditions, strategy characteristics, and money management also have influences on the profitability of the system trading. Unexpected price deviations from the predetermined trading rules can incur large losses to system traders. Therefore, most professional traders use strategy portfolios rather than only one strategy. Building a good strategy portfolio is important because trading performance depends on strategy portfolios. Despite of the importance of designing strategy portfolio, rule of thumb methods have been used to select trading strategies. In this study, we propose a SVM-based strategy portfolio management system. SVM were introduced by Vapnik and is known to be effective for data mining area. It can build good portfolios within a very short period of time. Since SVM minimizes structural risks, it is best suitable for the futures trading market in which prices do not move exactly the same as the past. Our system trading strategies include moving-average cross system, MACD cross system, trend-following system, buy dips and sell rallies system, DMI system, Keltner channel system, Bollinger Bands system, and Fibonacci system. These strategies are well known and frequently being used by many professional traders. We program these strategies for generating automated system signals for entry and exit. We propose SVM-based strategies selection system and portfolio construction and order routing system. Strategies selection system is a portfolio training system. It generates training data and makes SVM model using optimal portfolio. We make $m{\times}n$ data matrix by dividing KOSPI 200 index futures data with a same period. Optimal strategy portfolio is derived from analyzing each strategy performance. SVM model is generated based on this data and optimal strategy portfolio. We use 80% of the data for training and the remaining 20% is used for testing the strategy. For training, we select two strategies which show the highest profit in the next day. Selection method 1 selects two strategies and method 2 selects maximum two strategies which show profit more than 0.1 point. We use one-against-all method which has fast processing time. We analyse the daily data of KOSPI 200 index futures contracts from January 1990 to November 2011. Price change rates for 50 days are used as SVM input data. The training period is from January 1990 to March 2007 and the test period is from March 2007 to November 2011. We suggest three benchmark strategies portfolio. BM1 holds two contracts of KOSPI 200 index futures for testing period. BM2 is constructed as two strategies which show the largest cumulative profit during 30 days before testing starts. BM3 has two strategies which show best profits during testing period. Trading cost include brokerage commission cost and slippage cost. The proposed strategy portfolio management system shows profit more than double of the benchmark portfolios. BM1 shows 103.44 point profit, BM2 shows 488.61 point profit, and BM3 shows 502.41 point profit after deducting trading cost. The best benchmark is the portfolio of the two best profit strategies during the test period. The proposed system 1 shows 706.22 point profit and proposed system 2 shows 768.95 point profit after deducting trading cost. The equity curves for the entire period show stable pattern. With higher profit, this suggests a good trading direction for system traders. We can make more stable and more profitable portfolios if we add money management module to the system.

• Proceedings of the Membrane Society of Korea Conference
• /
• 1999.07a
• /
• pp.39-42
• /
• 1999

Perfluoropolymers represent the ultimate resistance to hostile chemical environments and high service temperature, attributed to the presence of fluorine in the polymer backbone, i.e. to the high bond energy of C-F and C-C bonds of fluorocarbons. Copolymers of Tetrafluoroethylene (TEE) and 2, 2, 4Trifluoro-5Trifluorometoxy- 1, 3Dioxole (TTD), commercially known as HYFLON AD, are amorphous perfluoropolymers with glass transition temperature (Tg)higher than room temperature, showing a thermal decomposition temperature exceeding 40$0^{\circ}C$. These polymer systems are highly soluble in fluorinated solvents, with low solution viscosities. This property allows the preparation of self-supported and composite membranes with desired membrane thickness. Symmetric and asymmetric perfluoropolymer membranes, made with HYFLON AD, have been prepared and evaluated. Porous and not porous symmetric membranes have been obtained by solvent evaporation with various processing conditions. Asymmetric membranes have been prepared by th wet phase inversion method. Measure of contact angle to distilled water have been carried out. Figure 1 compares experimental results with those of other commercial membranes. Contact angles of about 120$^{\circ}$for our amorphous perfluoropolymer membranes demonstrate that they posses a high hydrophobic character. Measure of contact angles to hexandecane have been also carried out to evaluate the organophobic character. Rsults are reported in Figure 2. The observed strong organophobicity leads to excellent fouling resistance and inertness. Porous membranes with pore size between 30 and 80 nanometers have shown no permeation to water at pressures as high as 10 bars. However high permeation to gases, such as O2, N2 and CO2, and no selectivities were observed. Considering the porous structure of the membrane, this behavior was expected. In consideration of the above properties, possible useful uses in th field of gas- liquid separations are envisaged for these membranes. A particularly promising application is in the field of membrane contactors, equipments in which membranes are used to improve mass transfer coefficients in respect to traditional extraction and absorption processes. Gas permeation properties have been evaluated for asymmetric membranes and composite symmetric ones. Experimental permselectivity values, obtained at different pressure differences, to various single gases are reported in Tab. 1, 2 and 3. Experimental data have been compared with literature data obtained with membranes made with different amorphous perfluoropolymer systems, such as copolymers of Perfluoro2, 2dimethyl dioxole (PDD) and Tetrafluorethylene, commercialized by the Du Pont Company with the trade name of Teflon AF. An interesting linear relationship between permeability and the glass transition temperature of the polymer constituting the membrane has been observed. Results are descussed in terms of polymer chain structure, which affects the presence of voids at molecular scale and their size distribution. Molecular Dyanmics studies are in progress in order to support the understanding of these results. A modified Theodoru- Suter method provided by the Amorphous Cell module of InsightII/Discover was used to determine the chain packing. A completely amorphous polymer box of about 3.5 nm was considered. Last but not least the use of amorphous perfluoropolymer membranes appears to be ideal when separation processes have to be performed in hostile environments, i.e. high temperatures and aggressive non-aqueous media, such as chemicals and solvents. In these cases Hyflon AD membranes can exploit the outstanding resistance of perfluoropolymers.

• Journal of vocational education research
• /
• v.35 no.4
• /
• pp.131-153
• /
• 2016

The purpose of this study was to derive evaluator's role and capability for institution accreditation evaluation of NCS-based vocational competency development training. This study attempted to explore in various ways evaluator's minute roles using Delphi method, and to derive knowledge, skill, attitude and integrity needed to verify the validity. To the end, this study conducted the Delphi research for over three rounds by selecting education training professionals and review evaluation professions as professional panels. From the results, roles of evaluators were defined as the total eight items including operator, moderator-mediator, cooperator, analyzer, verifier, institution evaluator, institution consultant, and learner, and the derived capabilities with respect to each role were 25 items in total. The area of knowledge included four items of capabilities such as HRD knowledge, NCS knowledge, knowledge of vocational competency development training, and knowledge of training institution accreditation evaluation, and the area of skill comprised fourteen items of capabilities such as conflict management ability, interpersonal relation ability, word processing ability, problem-solving ability, analysis ability, pre-preparation ability, time management ability, decision making ability, information comprehension and utilization ability, comprehensive thinking ability, understanding ability of vocational competency development training institutions, communication ability, feedback ability, and core understanding ability. The area of attitude was summarized with the seven items in total including subjectivity and fairness, service mind, sense of calling, ethics, self-development, responsibility, and teamwork. The knowledge, skill and attitude derived from the results of this study may be utilized to design and provide education programs conducive to qualitative and systematic accreditation and assessment to evaluators equipped with essential prerequisites. It is finally expected that this study will be helpful for designing module education programs by ability and for managing evaluator's quality in order to perform pre-service education and in-service education according to evaluator's experience and role.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.1
• /
• pp.9-18
• /
• 2024

The importance of active sonar systems is emerging due to the quietness of underwater targets and the increase in ambient noise due to the increase in maritime traffic. However, the low signal-to-noise ratio of the echo signal due to multipath propagation of the signal, various clutter, ambient noise and reverberation makes it difficult to identify underwater targets using active sonar. Attempts have been made to apply data-based methods such as machine learning or deep learning to improve the performance of underwater target recognition systems, but it is difficult to collect enough data for training due to the nature of sonar datasets. Methods based on mathematical modeling have been mainly used to compensate for insufficient active sonar data. However, methodologies based on mathematical modeling have limitations in accurately simulating complex underwater phenomena. Therefore, in this paper, we propose a sonar signal synthesis method based on a deep neural network. In order to apply the neural network model to the field of sonar signal synthesis, the proposed method appropriately corrects the attention-based encoder and decoder to the sonar signal, which is the main module of the Tacotron model mainly used in the field of speech synthesis. It is possible to synthesize a signal more similar to the actual signal by training the proposed model using the dataset collected by arranging a simulated target in an actual marine environment. In order to verify the performance of the proposed method, Perceptual evaluation of audio quality test was conducted and within score difference -2.3 was shown compared to actual signal in a total of four different environments. These results prove that the active sonar signal generated by the proposed method approximates the actual signal.