• Korean Journal of Remote Sensing
• /
• v.37 no.5_2
• /
• pp.1307-1315
• /
• 2021

The Changjiang Diluted Water (CDW) spreads over the East China Sea every summer and significantly affects the sea surface salinity changes in the seas around Jeju Island and the southern coast of Korea peninsula. Sometimes its effect extends to the eastern coast of Korea peninsula through the Korea Strait. Specifically, the CDW has a significant impact on marine physics and ecology and causes damage to fisheries and aquaculture. However, due to the limited field surveys, continuous observation of the CDW in the East China Sea is practically difficult. Many studies have been conducted using satellite measurements to monitor CDW distribution in near-real time. In this study, an algorithm for estimating Sea Surface Salinity (SSS) in the East China Sea was developed using the Geostationary Ocean Color Imager (GOCI). The Multilayer Perceptron Neural Network (MPNN) method was employed for developing an algorithm, and Soil Moisture Active Passive (SMAP) SSS data was selected for the output. In the previous study, an algorithm for estimating SSS using GOCI was trained by 2016 observation data. By comparison, the train data period was extended from 2015 to 2020 to improve the algorithm performance. The validation results with the National Institute of Fisheries Science (NIFS) serial oceanographic observation data from 2011 to 2019 show 0.61 of coefficient of determination (R²) and 1.08 psu of Root Mean Square Errors (RMSE). This study was carried out to develop an algorithm for monitoring the surface salinity of the East China Sea using GOCI and is expected to contribute to the development of the algorithm for estimating SSS by using GOCI-II.

• Journal of Korea Water Resources Association
• /
• v.51 no.12
• /
• pp.1217-1227
• /
• 2018

This paper aims to evaluate the applicability of dam inflow prediction model using recurrent neural network theory. To achieve this goal, the Artificial Neural Network (ANN) model and the Elman Recurrent Neural Network(RNN) model were applied to hydro-meteorological data sets for the Soyanggang dam and the Chungju dam basin during dam operation period. For the model training, inflow, rainfall, temperature, sunshine duration, wind speed were used as input data and daily inflow of dam for 10 days were used for output data. The verification was carried out through dam inflow prediction between July, 2016 and June, 2018. The results showed that there was no significant difference in prediction performance between ANN model and the Elman RNN model in the Soyanggang dam basin but the prediction results of the Elman RNN model are comparatively superior to those of the ANN model in the Chungju dam basin. Consequently, the Elman RNN prediction performance is expected to be similar to or better than the ANN model. The prediction performance of Elman RNN was notable during the low dam inflow period. The performance of the multiple hidden layer structure of Elman RNN looks more effective in prediction than that of a single hidden layer structure.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.6
• /
• pp.169-180
• /
• 2018

White light-emitting diodes (LEDs) are more economical than fluorescent lights, and provide high brightness, a high lifetime expectancy, and greater durability. As LEDs are closely connected with people's daily lives, dimming control of LED is an important component in providing energy savings and improving quality of life. In visible light communications systems using these LEDs, multiple input multiple output (MIMO) technology has attracted a lot of attention, in that it can attain the channel capacity in proportion to the number of antennas. This paper analyzes the power performance of three kinds of modulation in visible light communications (VLC) systems applied space-time block code (STBC) techniques. The modulation schemes are return-to-zero on-off keying (RZ-OOK), variable pulse position modulation (VPPM), and overlapping pulse position modulation (OPPM), and dimming control was applied. The power requirements and power consumption were used as metrics to compare the power efficiency in $2{\times}2$ STBC-VLC environments under the three kinds of modulation. We confirm that dimming control affects the communications performance of each modulation scheme. VPPM showed greater consumption among the three modulations, and OPPM showed energy savings comparable to VPPM.

• Journal of the Korean Orthopaedic Association
• /
• v.53 no.6
• /
• pp.466-477
• /
• 2018

Orthopaedics is an area where 3-dimensional (3D) printing technology is most likely to be utilized because it has been used to treat a range of diseases of the whole body. For arthritis, spinal diseases, trauma, deformities, and tumors, 3D printing can be used in the form of anatomical models, surgical guides, metal implants, bio-ceramic body reconstruction, and orthosis. In particular, in orthopaedic oncology, patients have a wide variety of tumor locations, but limited options for the limb salvage surgery have resulted in many complications. Currently, 3D printing personalized implants can be fabricated easily in a short time, and it is anticipated that all bone tumors in various surgical sites will be reconstructed properly. An improvement of 3D printing technology in the healthcare field requires close cooperation with many professionals in the design, printing, and validation processes. The government, which has determined that it can promote the development of 3D printing-related industries in other fields by leading the use of 3D printing in the medical field, is also actively supporting with an emphasis on promotion rather than regulation. In this review, the experience of using 3D printing technology for bone tumor surgery was shared, expecting orthopaedic surgeons to lead 3D printing in the medical field.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.112-119
• /
• 2019

In this paper, we propose a deep learning structure suitable for embedded system. The flame detection process of the proposed deep learning structure consists of four steps : flame area detection using flame color model, flame image classification using deep learning structure for flame color specialization, $N{\times}N$ cell separation in detected flame area, flame image classification using deep learning structure for flame shape specialization. First, only the color of the flame is extracted from the input image and then labeled to detect the flame area. Second, area of flame detected is the input of a deep learning structure specialized in flame color and is classified as flame image only if the probability of flame class at the output is greater than 75%. Third, divide the detected flame region of the images classified as flame images less than 75% in the preceding section into $N{\times}N$ units. Fourthly, small cells divided into $N{\times}N$ units are inserted into the input of a deep learning structure specialized to the shape of the flame and each cell is judged to be flame proof and classified as flame images if more than 50% of cells are classified as flame images. To verify the effectiveness of the proposed deep learning structure, we experimented with a flame database of ImageNet. Experimental results show that the proposed deep learning structure has an average resource occupancy rate of 29.86% and an 8 second fast flame detection time. The flame detection rate averaged 0.95% lower compared to the existing deep learning structure, but this was the result of light construction of the deep learning structure for application to embedded systems. Therefore, the deep learning structure for flame detection proposed in this paper has been proved suitable for the application of embedded system.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.3
• /
• pp.419-432
• /
• 2019

Most of deep learning model training was proceeded by supervised learning, which is to train labeling data composed by inputs and corresponding outputs. Labeling data was directly generated manually, so labeling accuracy of data is relatively high. However, it requires heavy efforts in securing data because of cost and time. Additionally, the main goal of supervised learning is to improve detection performance for 'True Positive' data but not to reduce occurrence of 'False Positive' data. In this paper, the occurrence of unpredictable 'False Positive' appears by trained modes with labeling data and 'True Positive' data in monitoring of deep learning-based CCTV accident detection system, which is under operation at a tunnel monitoring center. Those types of 'False Positive' to 'fire' or 'person' objects were frequently taking place for lights of working vehicle, reflecting sunlight at tunnel entrance, long black feature which occurs to the part of lane or car, etc. To solve this problem, a deep learning model was developed by simultaneously training the 'False Positive' data generated in the field and the labeling data. As a result, in comparison with the model that was trained only by the existing labeling data, the re-inference performance with respect to the labeling data was improved. In addition, re-inference of the 'False Positive' data shows that the number of 'False Positive' for the persons were more reduced in case of training model including many 'False Positive' data. By training of the 'False Positive' data, the capability of field application of the deep learning model was improved automatically.

• Journal of the Korean Society of Radiology
• /
• v.14 no.7
• /
• pp.899-906
• /
• 2020

The WSS lateral examination is important for diagnosing spinal disorders. Recently, long-length detectors for large-area diagnose have been popularized to effectively reduce the exposure dose and examination time. It can be applied very efficiently to examinations of patients with high risk of falls, children, and adolescents. However, since the image is acquired through a single irradiation, the volume of cervical vertebra is relatively smaller than the lumbar due to the geometrical anatomy of the spine. Therefore, this study intends to fabricate an additional filter using 3D printing technology and copper filament to obtain uniform image quality in the WSS lateral examination and to analyze the results. 3D printing technology is able to easily print a desired shape, so it is widely used in the entire industrial field, and recently, a copper filament has been developed to confirm the possibility as an additional filter. In the WSS lateral examination, CNR and SNR were excellently measured when the additional filter was applied, confirming the possibility of using the additional filter.

• Journal of Internet Computing and Services
• /
• v.21 no.6
• /
• pp.1-12
• /
• 2020

Low-Orbit satellite mobile networks can provide services through miniaturized terminals with low transmission power, which can be used as reliable means of communication in the national public disaster network and defense sector. However, the high traffic environment in the emergency preparedness situation increases the new call blocking probability and the handover failure probability of the satellite network, and the increase of the handover failure probability affects the QoS because low orbit satellites move in orbit at a very high speed. Among the channel allocation methods of satellite communication, the FCA shows relatively better performance in a high traffic environment than DCA and is suitable for emergency preparedness situations, but in order to optimize QoS when traffic increases, the new call blocking and the handover failure must be minimized. In this paper, we propose LEO-DBC (LEO satellite dynamic beam width control) technique, which improves QoS by adaptive adjustment of beam width of low-orbit satellites and call time of terminals by improving FCA-QH method. Through the LEO-DBC technique, it is expected that the QoS of the mobile satellite communication network can be optimally maintained in high traffic environments in emergency preparedness situations.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.1
• /
• pp.409-414
• /
• 2021

Tightness performance that blocks compartments is important for surface ships to achieve superior mission performance and survivability in combat environments. To meet the above requirements, airtightness of the structural elements and the appropriate strength to specific areas are checked during a test run after ship construction. In particular, air tests of compartments adjacent to the water surface are performed. In an air test, air is injected into the compartment up to the test pressure of the test memo. The pressure drop value is checked after 10 minutes to determine if the requirements of the corresponding area are satisfied. In summer, however, when the influence of the outside temperature is large, a phenomenon in which the internal pressure increases during the air test was identified. This phenomenon reduces the reliability of the test result. Therefore, a system was designed to compensate for temperature changes in the compartments through this study. The developed system calculates the amount of pressure change caused by a temperature change in the compartment and outputs a correction value. The pressure change was calculated using the ideal gas equation, reflecting the maintenance, increase, and decrease in temperature during the test process. A comparison of the calculated pressure correction value with the database of NIST REFPROP revealed a difference of 0.126% to a maximum of 0.253%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.12
• /
• pp.1711-1717
• /
• 2020

X-ray of up to 9MeV are used for container inspection. X-ray intensity must be maintained stably regardless of changes in time. If dose is not constant, it may affect the image quality, and as a result, may affect the inspection of abnormal cargo. Therefore, to acquire high-quality images, continuous dose monitoring is required. In this study, the ion-chamber based device was developed for monitoring the dose change in high-energy x-ray. And to estimate the performance of signal-processing device change according to the environmental change, the output changing due to the change of temperature and humidity was observed. In addition, verification of the device was performed by measuring the output change. As a result of the measurement, there was no significant difference in performance due to changes in temperature and humidity, and the change in output according to the change in exposure was linear. Therefore, it was found that the developed device is suitable for the dose monitoring of high-energy x-ray.