• Korean Journal of Agricultural and Forest Meteorology
• /
• v.19 no.3
• /
• pp.153-163
• /
• 2017

The objective of this study is to enhance the model's speed of estimating weather variables (e.g., minimum/maximum temperature, sunshine hour, PRISM (Parameter-elevation Regression on Independent Slopes Model) based precipitation), which are applied to the Agrometeorological Early Warning System (http://www.agmet.kr). The current process of weather estimation is operated on high-performance multi-core CPUs that have 8 physical cores and 16 logical threads. Nonetheless, the server is not even dedicated to the handling of a single county, indicating that very high overhead is involved in calculating the 10 counties of the Seomjin River Basin. In order to reduce such overhead, several cache and parallelization techniques were used to measure the performance and to check the applicability. Results are as follows: (1) for simple calculations such as Growing Degree Days accumulation, the time required for Input and Output (I/O) is significantly greater than that for calculation, suggesting the need of a technique which reduces disk I/O bottlenecks; (2) when there are many I/O, it is advantageous to distribute them on several servers. However, each server must have a cache for input data so that it does not compete for the same resource; and (3) GPU-based parallel processing method is most suitable for models such as PRISM with large computation loads.

• Journal of the Korean Society of Radiology
• /
• v.6 no.6
• /
• pp.447-454
• /
• 2012

To examine the performance of a diagnostic X-ray system, we tested a linearity, reproducibility, and Half Value Layer(HVL). The linearity was examined 4 times of irradiation with a given condition, and we recorded a level of radiation. We then calculated the mR/mAs. And the measured value should not be more than 0.1. If the measured value was more than 0.1, we could know that the linearity was decreased. The reproducibility was analyzed 10 times of irradiations at 80kVp, 200mA, 20mAs and 120kVp, 300mA, 8mAs. The values from these analyses were integrated into CV equation, and we could get outputs. The reproducibility was good if the output was lower than 0.05. HVL was measured 3 times of irradiation without a filter, and we inserted additional HLV filters with 0, 1, 2, 4 mm of thickness. We tested the values until we get the measured value less than a half of the value measured without additional filter. We tested the linearity, the reproducibility, and HVL of 5 diagnostic X-ray generators in this facilities. The linearity of No. 1 and No. 5 generator didn't satisfy the standard for radiation safety around 300mA~400mA and 100mA~200mA, respectively. HVL of No.1 generator was not satisfied at 80kVp. The outputs were higher in the three-phase equipment than the single-phase equipment. The old generators need to maintain and exchange of components based on the these results. Then, we could contribute to getting more exact diagnosis increasing a quality of the image and decreasing an expose dose of radiation.

• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.4
• /
• pp.15-24
• /
• 2007

The recognition of postal address images is indispensable for the automatic sorting of postal envelopes. The process of the address image recognition is composed of three steps-address image preprocessing, character recognition, address interpretation. The extracted character images from the preprocessing step are forwarded to the character recognition step, in which multiple candidate characters with reliability scores are obtained for each character image extracted. aracters with reliability scores are obtained for each character image extracted. Utilizing those character candidates with scores, we obtain the final valid address for the input envelope image through the address interpretation step. The envelope sorting rate depends on the performance of all three steps, among which character recognition step could be said to be very important. The good character recognizer would be the one which could produce valid candidates with very reliable scores to help the address interpretation step go easy. In this paper, we propose the method of generating character candidates with reliable recognition scores. We utilize the existing MLP(multilayered perceptrons) neural network of the address recognition system in the current automatic postal envelope sorters, as the classifier for the each image from the preprocessing step. The MLP is well known to be one of the best classifiers in terms of processing speed and recognition rate. The false alarm problem, however, might be occurred in recognition results, which made the address interpretation hard. To make address interpretation easy and improve the envelope sorting rate, we propose promising methods to reestimate the recognition score (confidence) of the existing MLP classifier: the generation method of the statistical recognition properties of the classifier and the method of the combination of the MLP and the subspace classifier which roles as a reestimator of the confidence. To confirm the superiority of the proposed method, we have used the character images of the real postal envelopes from the sorters in the post office. The experimental results show that the proposed method produces high reliability in terms of error and rejection for individual characters and non-characters.

• Journal of the Korean Electrochemical Society
• /
• v.10 no.4
• /
• pp.283-287
• /
• 2007

KIER has been developing the anode-supported flat tubular solid oxide fuel cell unit bundle for the intermediate temperature($700{\sim}800^{\circ}C$) operation. Anode-supported flat tubular cells have Ni/YSZ cermet anode support, 8 moi.% $Y_2O_3$ stabilized $ZrO_2(YSZ)$ thin electrolyte, and cathode multi-layer composed of Sr-doped $LaSrMnO_3(LSM)$, LSM-YSZ composite, and $LaSrCoFeO_3(LSCF)$. The prepared anode-supported flat tubular cell was joined with ferritic stainless steel cap by induction brazing process. Current collection for the cathode was achieved by winding Ag wire and $La_{0.6}Sr_{0.4}CoO_3(LSCo)$ paste, while current collection for the anode was achieved by using Ni wire and felt. For making stack, the prepared anode-supported flat tubular cells with effective electrode area of $90\;cm^2$ connected in series with 12 unit bundles, in which unit bundle consists of two cells connected in parallel. The performance of unit bundle in 3% humidified $H_2$ and air at $800^{\circ}C$ shows maximum power density of $0.39\;W/cm^2$ (@ 0.7V). Through these experiments, we obtained basic technology of the anode-supported flat tubular cell and established the proprietary concept of the anode-supported flat tubular cell unit bundle.

• Journal of IKEEE
• /
• v.25 no.3
• /
• pp.493-500
• /
• 2021

In this paper, we propose the development of deep learning structure to improve quality of polygonal containers. The deep learning structure consists of a convolution layer, a bottleneck layer, a fully connect layer, and a softmax layer. The convolution layer is a layer that obtains a feature image by performing a convolution 3x3 operation on the input image or the feature image of the previous layer with several feature filters. The bottleneck layer selects only the optimal features among the features on the feature image extracted through the convolution layer, reduces the channel to a convolution 1x1 ReLU, and performs a convolution 3x3 ReLU. The global average pooling operation performed after going through the bottleneck layer reduces the size of the feature image by selecting only the optimal features among the features of the feature image extracted through the convolution layer. The fully connect layer outputs the output data through 6 fully connect layers. The softmax layer multiplies and multiplies the value between the value of the input layer node and the target node to be calculated, and converts it into a value between 0 and 1 through an activation function. After the learning is completed, the recognition process classifies non-circular glass bottles by performing image acquisition using a camera, measuring position detection, and non-circular glass bottle classification using deep learning as in the learning process. In order to evaluate the performance of the deep learning structure to improve quality of polygonal containers, as a result of an experiment at an authorized testing institute, it was calculated to be at the same level as the world's highest level with 99% good/defective discrimination accuracy. Inspection time averaged 1.7 seconds, which was calculated within the operating time standards of production processes using non-circular machine vision systems. Therefore, the effectiveness of the performance of the deep learning structure to improve quality of polygonal containers proposed in this paper was proven.

• Journal of Internet Computing and Services
• /
• v.21 no.6
• /
• pp.23-31
• /
• 2020

This study was carried out to generate various images of railroad surfaces with random defects as training data to be better at the detection of defects. Defects on the surface of railroads are caused by various factors such as friction between track binding devices and adjacent tracks and can cause accidents such as broken rails, so railroad maintenance for defects is necessary. Therefore, various researches on defect detection and inspection using image processing or machine learning on railway surface images have been conducted to automate railroad inspection and to reduce railroad maintenance costs. In general, the performance of the image processing analysis method and machine learning technology is affected by the quantity and quality of data. For this reason, some researches require specific devices or vehicles to acquire images of the track surface at regular intervals to obtain a database of various railway surface images. On the contrary, in this study, in order to reduce and improve the operating cost of image acquisition, we constructed the 'Defective Railroad Surface Regeneration Model' by applying the methods presented in the related studies of the Generative Adversarial Network (GAN). Thus, we aimed to detect defects on railroad surface even without a dedicated database. This constructed model is designed to learn to generate the railroad surface combining the different railroad surface textures and the original surface, considering the ground truth of the railroad defects. The generated images of the railroad surface were used as training data in defect detection network, which is based on Fully Convolutional Network (FCN). To validate its performance, we clustered and divided the railroad data into three subsets, one subset as original railroad texture images and the remaining two subsets as another railroad surface texture images. In the first experiment, we used only original texture images for training sets in the defect detection model. And in the second experiment, we trained the generated images that were generated by combining the original images with a few railroad textures of the other images. Each defect detection model was evaluated in terms of 'intersection of union(IoU)' and F1-score measures with ground truths. As a result, the scores increased by about 10~15% when the generated images were used, compared to the case that only the original images were used. This proves that it is possible to detect defects by using the existing data and a few different texture images, even for the railroad surface images in which dedicated training database is not constructed.

• Korean Chemical Engineering Research
• /
• v.60 no.4
• /
• pp.588-593
• /
• 2022

Terephthalaldehyde (TPA) is introduced as a cross liker to enhance electron transfer of hemin-based cathodic catalyst consisting of polyethyleneimine (PEI), carbon nanotube (CNT) for hydrogen peroxide reduction reaction (HPRR). In the cyclic voltammetry (CV) test with 10 mM H₂O₂ in phosphate buffer solution (pH 7.4), the current density for HPRR of the suggested catalyst (CNT/PEI/hemin/PEI/TPA) shows 0.2813 mA cm^-2 (at 0.2 V vs. Ag/AgCl), which is 2.43 and 1.87 times of non-cross-linked (CNT/PEI/hemin/PEI) and conventional cross liker (glutaraldehyde, GA) used catalyst (CNT/PEI/hemin/PEI/GA), respectively. In the case of onset potential for HPRR, that of CNT/PEI/hemin/PEI/TPA is observed at 0.544 V, while those of CNT/PEI/hemin/PEI and CNT/PEI/hemin/PEI/GA are 0.511 and 0.471 V, respectively. These results indicate that TPA plays a role in facilitating electron transfer between the electrodes and substrates due to the π-conjugated cross-linking bonds, whereas conventional GA cross-linker increases the overpotential by interrupting electron and mass transfer. Electrochemical impedance spectroscopy (EIS) results also display the same tendency. The charge transfer resistance (R_ct) of CNT/PEI/hemin/PEI/TPA decreases about 6.2% from that of CNT/PEI/hemin/PEI, while CNT/PEI/hemin/PEI/GA shows the highest R_ct. The polarization curve using each catalyst also supports the superiority of TPA cross liker. The maximum power density of CNT/PEI/hemin/PEI/TPA (36.34±1.41 μWcm^-2) is significantly higher than those of CNT/PEI/hemin/PEI (27.87±0.95 μWcm^-2) and CNT/PEI/hemin/PEI/GA (25.57±1.32 μWcm^-2), demonstrating again that the cathode using TPA has the best performance in HPRR.

• Journal of the Korean Society of Radiology
• /
• v.5 no.5
• /
• pp.303-308
• /
• 2011

Background/Aims : The increasing use of diagnostic and therapeutic interventional radiology calls for greater consideration of radiation exposure risk to radiologist and radiological technician, and emphasizes the proper system of radiation protection. This study was designed to assess the effect of a protective grass shield. Methods : A protective grass was following data depth, 0.8 cm; width, 100 cm; length, 100 cm, lead equivalent, 1.6 mmPb. The protective shield was located between the patient and the radiologist. Thirty patients (13 male and 17 female) undergoing interventional radiology between September 2010 and December 2010 were selected for this study. The dose of radiation exposure was recorded with or without the protective grass shield at the level of the head, chest, and pelvis. The measurement was made at 50 cm and 150 cm from the radiation source. Results : The mean patient age was 69 years. The mean patient height and weight was $159.7{\pm}6.7$ cm and $60.3{\pm}5.9$ kg, respectively. The mean body mass index (BMI) was $20.5{\pm}3.0$ kg/m2. radiologists received $1530.2{\pm}550.0$ mR/hr without the protective lead shield. At the same distance, radiation exposure was significantly reduced to $50.3{\pm}85.2$ mR/hr with the protective lead shield (p-value<0.0001). The radiation exposure to radiologist and radiological technician was significantly reduced by the use of a protective lead shield (p value <0.0001). The amount of radiation exposure during interventional radiology was related to the patient' BMI (r=0.749, p=0.001). Conclusions : This protective shield grass is effective in protecting radiologist and radiological technician from radiation exposure.

• Progress in Medical Physics
• /
• v.21 no.1
• /
• pp.113-119
• /
• 2010

Software for GafChromic EBT2 film dosimetry was developed in this study. The software provides film calibration functions based on color channels, which are categorized depending on the colors red, green, blue, and gray. Evaluations of the correction effects for light scattering of a flat-bed scanner and thickness differences of the active layer are available. Dosimetric results from EBT2 films can be compared with those from the treatment planning system ECLIPSE or the two-dimensional ionization chamber array MatriXX. Dose verification using EBT2 films is implemented by carrying out the following procedures: file import, noise filtering, background correction and active layer correction, dose calculation, and evaluation. The relative and absolute background corrections are selectively applied. The calibration results and fitting equation for the sensitometric curve are exported to files. After two different types of dose matrixes are aligned through the interpolation of spatial pixel spacing, interactive translation, and rotation, profiles and isodose curves are compared. In addition, the gamma index and gamma histogram are analyzed according to the determined criteria of distance-to-agreement and dose difference. The performance evaluations were achieved by dose verification in the $60^{\circ}$-enhanced dynamic wedged field and intensity-modulated (IM) beams for prostate cancer. All pass ratios for the two types of tests showed more than 99% in the evaluation, and a gamma histogram with 3 mm and 3% criteria was used. The software was developed for use in routine periodic quality assurance and complex IM beam verification. It can also be used as a dedicated radiochromic film software tool for analyzing dose distribution.

• Journal of Biomedical Engineering Research
• /
• v.19 no.5
• /
• pp.455-468
• /
• 1998

For the objective interpretation of cerebral metabolic patterns in epilepsy patients, we developed computer-aided classifier using artificial neural network. We studied interictal brain FDG PET scans of 257 epilepsy patients who were diagnosed as normal(n=64), L TLE (n=112), or R TLE (n=81) by visual interpretation. Automatically segmented volume of interest (VOI) was used to reliably extract the features representing patterns of cerebral metabolism. All images were spatially normalized to MNI standard PET template and smoothed with 16mm FWHM Gaussian kernel using SPM96. Mean count in cerebral region was normalized. The VOls for 34 cerebral regions were previously defined on the standard template and 17 different counts of mirrored regions to hemispheric midline were extracted from spatially normalized images. A three-layer feed-forward error back-propagation neural network classifier with 7 input nodes and 3 output nodes was used. The network was trained to interpret metabolic patterns and produce identical diagnoses with those of expert viewers. The performance of the neural network was optimized by testing with 5~40 nodes in hidden layer. Randomly selected 40 images from each group were used to train the network and the remainders were used to test the learned network. The optimized neural network gave a maximum agreement rate of 80.3% with expert viewers. It used 20 hidden nodes and was trained for 1508 epochs. Also, neural network gave agreement rates of 75~80% with 10 or 30 nodes in hidden layer. We conclude that artificial neural network performed as well as human experts and could be potentially useful as clinical decision support tool for the localization of epileptogenic zones.