• Journal of Wetlands Research
• /
• v.25 no.3
• /
• pp.167-175
• /
• 2023

Precipitation data is one of the essential input datasets used in various fields such as wetland management, hydrological simulation, and water resource management. In order to efficiently manage water resources using precipitation data, it is essential to secure as much data as possible by minimizing the missing rate of data. In addition, more efficient hydrological simulation is possible if precipitation data for ungauged areas are secured. However, missing precipitation data have been estimated mainly by statistical equations. The purpose of this study is to propose a new method to restore missing precipitation data using machine learning algorithms that can predict new data based on correlations between data. Moreover, compared to existing statistical methods, the applicability of machine learning techniques for restoring missing precipitation data is evaluated. Representative machine learning algorithms, Artificial Neural Network (ANN) and Random Forest (RF), were applied. For the performance of classifying the occurrence of precipitation, the RF algorithm has higher accuracy in classifying the occurrence of precipitation than the ANN algorithm. The F1-score and Accuracy values, which are evaluation indicators of the classification model, were calculated as 0.80 and 0.77, while the ANN was calculated as 0.76 and 0.71. In addition, the performance of estimating precipitation also showed higher accuracy in RF than in ANN algorithm. The RMSE of the RF and ANN algorithms was 2.8 mm/day and 2.9 mm/day, and the values were calculated as 0.68 and 0.73.

• Geophysics and Geophysical Exploration
• /
• v.13 no.4
• /
• pp.315-322
• /
• 2010

The phase-screen and the split-step Fourier migrations, which are implemented in both the frequency-wavenumber and frequency-space domains by using one-way scalar wave equation, allow imaging in laterally heterogeneous media with less computing time and efficiency. The generalized-screen migration employs the series expansion of the exponential, unlike the phase-screen and the split-step Fourier migrations which assume the vertical propagation in frequency-wavenumber domain. In addition, since the generalized-screen migration generalizes the series expansion of the vertical slowness, it can utilize higher-order terms of that series expansion. As a result, the generalized-screen migration has higher accuracy in computing the propagation with wide angles than the phase-screen and split-step Fourier migrations for media with large and rapid lateral velocity variations. In this study, we developed a 2D prestack generalized-screen migration module for imaging a complex subsurface efficiently, which includes various dips and large lateral variations. We compared the generalized-screen propagator with the phase-screen propagator for a constant perturbation model and the SEG/EAGE salt dome model. The generalized-screen propagator was more accurate than the phase-screen propagator in computing the propagation with wide angles. Furthermore, the more the higher-order terms were added for the generalized-screen propagator, the more the accuracy was increased. Finally, we compared the results of the generalizedscreen migration with those of the phase-screen migration for a model which included various dips and large lateral velocity variations and the synthetic data of the SEG/EAGE salt dome model. In the generalized-screen migration section, reflectors were positioned more accurately than in the phase-screen migration section.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.59 no.3
• /
• pp.29-39
• /
• 2017

In this study, the GPM (Global Precipitation Mission) IMERG (Integrated Multi-satellitE retrievals for GPM) rainfall data was verified and evaluated using ground AWS (Automated Weather Station) and radar in order to investigate the availability of GPM IMERG rainfall data. The SPI (Standardized Precipitation Index) was calculated based on the GPM IMERG data and also compared with the results obtained from the ground observation data for the Hoengseong Dam and Yongdam Dam areas. For the radar data, 1.5 km CAPPI rainfall data with a resolution of 10 km and 30 minutes was generated by applying the Z-R relationship ($Z=200R^{1.6}$) and used for accuracy verification. In order to calculate the SPI, PERSIANN_CDR and TRMM 3B42 were used for the period prior to the GPM IMERG data availability range. As a result of latency verification, it was confirmed that the performance is relatively higher than that of the early run mode in the late run mode. The GPM IMERG rainfall data has a high accuracy for 20 mm/h or more rainfall as a result of the comparison with the ground rainfall data. The analysis of the time scale of the SPI based on GPM IMERG and changes in normal annual precipitation adequately showed the effect of short term rainfall cases on local drought relief. In addition, the correlation coefficient and the determination coefficient were 0.83, 0.914, 0.689 and 0.835, respectively, between the SPI based GPM IMERG and the ground observation data. Therefore, it can be used as a predictive factor through the time series prediction model. We confirmed the hydrological utilization and the possibility of real time drought monitoring using SPI based on GPM IMERG rainfall, even though results presented in this study were limited to some rainfall cases.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.5B
• /
• pp.326-333
• /
• 2008

In wireless sensor networks, a sensor node with in-network aggregation adjusts the timeout which is a waiting time to receive messages from child nodes. This paper proposes a novel timeout scheduling scheme for data aggregation in wireless sensor networks, which adaptively configures its timeout according to changing data patterns in order to improve energy efficiency and data accuracy during data aggregation. The proposed scheme decreases the timeout when the temporal difference of collected data(data variation) from children is lower than a pre-defined threshold because there is no occurrence of critical events. Conversely, it increases the timeout when the data variation is higher than the pre-defined threshold in order to fulfill more accurate data aggregation. Extensive simulation reveals that the proposed scheme outperforms the cascading-based scheme in terms of energy consumption and data accuracy.

• Journal of Broadcast Engineering
• /
• v.23 no.2
• /
• pp.197-205
• /
• 2018

Panoramic image is one of the technologies that are commonly used today. However, technical difficulties still exist in panoramic video production. Without a special camera such as a 360-degree camera, making panoramic video becomes more difficult. In order to make a panoramic video, it is necessary to synchronize the timeline of multiple videos shot at multiple locations. However, the timeline synchronization method using the internal clock of the camera may cause an error due to the difference of the internal hardware. In order to solve this problem, timeline synchronization between multiple videos using visual information or auditory information has been studied. However, there is a problem in accuracy and processing time when using video information, and there is a problem in that, when using audio information, there is no synchronization when there is sensitivity to noise or there is no melody. Therefore, in this paper, we propose a timeline synchronization method between multiple video using audio waveform. It shows higher synchronization accuracy and temporal efficiency than the video information based time synchronization method.

• Journal of Internet of Things and Convergence
• /
• v.8 no.3
• /
• pp.55-61
• /
• 2022

In order to reduce greenhouse gases, the main culprit of global warming, the United Nations signed the Climate Change Convention in 1992. Korea is also pursuing a policy to expand the supply of renewable energy to reduce greenhouse gas emissions. The expansion of renewable energy development using solar power led to the expansion of wind power and solar power generation. The expansion of renewable energy development, which is greatly affected by weather conditions, is creating difficulties in managing the supply and demand of the power system. To solve this problem, the power brokerage market was introduced. Therefore, in order to participate in the power brokerage market, it is necessary to predict the amount of power generation. In this paper, the prediction system was used to analyze the Yonchuk solar power plant. As a result of applying solar insolation from on-site (Model 1) and the Korea Meteorological Administration (Model 2), it was confirmed that accuracy of Model 2 was 3% higher. As a result of comparative analysis of the DNN and RNN models, it was confirmed that the prediction accuracy of the DNN model improved by 1.72%.

• Korean Journal of Construction Engineering and Management
• /
• v.21 no.6
• /
• pp.113-124
• /
• 2020

Social movements to improve the performance of buildings through remodeling of aging apartment houses are being captured. To this end, the remodeling construction cost analysis, structural analysis, and political institutional review have been conducted to suggest ways to activate the remodeling. However, although the method of analyzing construction cost for remodeling apartment houses is currently being proposed for research purposes, there are limitations in practical application possibilities. Specifically, In order to be used practically, it is applicable to cases that have already been completed or in progress, but cases that will occur in the future are also used for construction cost analysis, so the sustainability of the analysis method is lacking. For the purpose of this, we would like to suggest an automated estimating method. For the sustainability of construction cost estimates, Deep-Learning was introduced in the estimating procedure. Specifically, a method for automatically finding the relationship between design elements, work types, and cost increase factors that can occur in apartment remodeling was presented. In addition, Monte Carlo Simulation was included in the estimation procedure to compensate for the lack of uncertainty, which is the inherent limitation of the Deep Learning-based estimation. In order to present higher accuracy as cases are accumulated, a method of calculating higher accuracy by comparing the estimate result with the existing accumulated data was also suggested. In order to validate the sustainability of the automated estimates proposed in this study, 13 cases of learning procedures and an additional 2 cases of cumulative procedures were performed. As a result, a new construction cost estimating procedure was automatically presented that reflects the characteristics of the two additional projects. In this study, the method of estimate estimate was used using 15 cases, If the cases are accumulated and reflected, the effect of this study is expected to increase.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.12
• /
• pp.1234-1238
• /
• 2010

In recent, there have been done many studies on applications that monitor the information of mobile objects using Wireless Sensor Networks (WSN). A density query that finds out an area spread by density that a target object requires in the whole sensing field is a field of object monitoring applications. In this paper, we propose a novel homogeneous network-based in-network density query processing scheme that significantly reduces query processing costs and assures high accuracy. This scheme is based on the possibility-based expected region selection technique and the result compensation technique in order to enhance the accuracy of the density query and to minimize its energy consumption. To show the superiority of our proposed scheme, we compare it with the existing density query processing scheme. As a result, our proposed scheme reduces about 92% energy consumption for query processing, while its network lifetime increases compared to the existing scheme. In addition, the proposed scheme guarantees higher accuracy than the existing scheme in terms of the query result.

• Journal of Korean Society of Forest Science
• /
• v.109 no.1
• /
• pp.23-30
• /
• 2020

The use of increased computing power, machine learning, and deep learning techniques have dramatically increased in various sectors. In particular, image detection algorithms are broadly used in forestry and remote sensing areas to identify forest types and tree species. However, in South Korea, machine learning has rarely, if ever, been applied in forestry image detection, especially to classify tree species. This study integrates the application of machine learning and forest image detection; specifically, we compared the ability of two machine learning data collection methods, namely image data captured by forest experts (D1) and web-crawling (D2), to automate the classification of five trees species. In addition, two methods of characterization to train/test the system were investigated. The results indicated a significant difference in classification accuracy between D1 and D2: the classification accuracy of D1 was higher than that of D2. In order to increase the classification accuracy of D2, additional data filtering techniques were required to reduce the noise of uncensored image data.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.26 no.1
• /
• pp.48-57
• /
• 2016

Objectives: This study was conducted to evaluate the accuracy and precision of airborne metal analysis using polyvinyl chloride(PVC) membrane filter by pretreatment methods. Methods: A total of 75 spiked PVC samples for Cr, Fe and Mn ranged from 6 ug/sample to 40 ug/sample were used to evaluate recovery rates for three pretreatment methods: acid extraction, hot plate ashing and microwave digestion. For Mn, an additional 75 spiked mixed cellulose ester(MCE) membrane filters were analysed to compare the recovery rates of PVC samples. All samples were analysed with an inductively coupled plasma optical emission spectrometer(ICP-OES) and manganese samples were additionally analyzed by atomic absorption spectrometer(AAS). Results: The overall mean recovery rates of PVC samples for Cr, Fe and Mn were 90% or higher regardless of pretreatment methods, but there were statistically significant differences in recovery rates for Cr(p<0.05) and Mn(p<0.01) samples by pretreatment methods. The biases and the coefficient variations of PVC samples for three metals pretreated with three kinds of pretreatment methods ranged from 1.7% to 4.7% and from 1.6% to 6.5%, respectively. The manganese PVC samples pretreated by microwave digestion and analyzed with ICP-OES had the lowest bias at 1.9% and also showed lower bias than the bias for MCE samples, 2.7%. Conclusions: In order to accurately analyze the metals sampled with PVC membrane filters, microwave digestion and ICP-OES can be recommended.