• Geophysics and Geophysical Exploration
• /
• v.23 no.3
• /
• pp.149-156
• /
• 2020

The genesis of ore bodies is a very diverse and complex process, and the target depth of mineral exploration increases. These create a need for predictive mineral exploration, which may be facilitated by the advancement of machine learning and geological database. In this study, we confirm that the faults and igneous rocks distributions and magnetic data can be used as input data for potential mapping using deep neural networks. When the input data are constructed with faults, igneous rocks, and magnetic data, we can build a potential mapping model of the metal deposit that has a predictive accuracy greater than 0.9. If detailed geological and geophysical data are obtained, this approach can be applied to the potential mapping on a mine scale. In addition, we confirm that the magnetic data, which provide the distribution of the underground igneous rock, can supplement the limited information from the surface igneous rock distribution. Therefore, rather than simply integrating various data sets, it will be more important to integrate information considering the geological correlation to genesis of minerals.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.9
• /
• pp.110-117
• /
• 2016

Applying the contrast enhancement algorithms to luminance values of color images is a widely used approach to enhance the contrast of color images. The results obtained by this approach have reduced saturation compared with that of the original images in spite of contrast enhancement without color degradation. Applying the contrast enhancement algorithm to each channel of color images is another approach for the contrast enhancement of color images. This method produces improved images in terms of contrast and saturation while the hue of original images is changed. In this paper, main cause of color degradation is analyzed and then solving the problem based on the analysis. The channel adaptive contrast enhancement method considering characteristics of each channel is also proposed to deal with color degradation. As a result, the proposed method enhances the contrast and saturation simultaneously without color degradation. Experimental results show that the proposed method outperforms the conventional methods not only on subjective evaluation but on objective criteria.

• Journal of KIISE
• /
• v.44 no.7
• /
• pp.674-679
• /
• 2017

In big-data environments wherein large amounts of text documents are produced daily, titles are very important clues that enable a prompt catching of the key ideas in documents; however, titles are absent for numerous document types such as blog articles and social-media messages. In this paper, a title-generation model for which sequence-to-sequence RNNs with attention and copying mechanisms are employed is proposed. For the proposed model, input sentences are encoded based on bi-directional GRU (gated recurrent unit) networks, and the title words are generated through a decoding of the encoded sentences with keywords that are automatically selected from the input sentences. Regarding the experiments with 93631 training-data documents and 500 test-data documents, the attention-mechanism performances are more effective (ROUGE-1: 0.1935, ROUGE-2: 0.0364, ROUGE-L: 0.1555) than those of the copying mechanism; in addition, the qualitative-evaluation radiative performance of the former is higher.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.14 no.4
• /
• pp.269-276
• /
• 2012

The objective of this short communication is to introduce several evaluation methods to crop model users because the evaluation of crop model performance is an important step to develop or select crop model. In this paper, mean error, mean absolute error, index of agreement, root mean square error, efficiency of model, accuracy factor and bias factor were explained and compared in terms of dimension and observed number. Efficiency of model and index of agreement are dimensionless and independent of number of observation. Relative root mean square, accuracy factor and bias factor are dimensionless and not independent of number of observation. Mean error and mean absolute error are affected by dimension and number of observation.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.6D
• /
• pp.727-736
• /
• 2009

Many of the researchers have tried to enhance the satisfaction of both users and owners of social infrastructures by applying the asset management methodology. This study is to develop more efficient asset management framework for bridge management. Based on various literature review, an asset management procedure for bridge management was suggested, and appropriate practices at each step were given. The suggested procedures include the determination of operation and maintenance strategy, level of service, performance measure, valuation of assets, and decision-makings. In addition, this study suggests an applicable decision-making process for the resource distribution based on the management strategy.

• Journal of the Korean Society of Radiology
• /
• v.17 no.5
• /
• pp.761-766
• /
• 2023

In this paper, TI was measured to classify the presence or absence of mild left ventricular diastolic dysfunction. Of the total 306 data, 206 were used as training data and 100 were used as test data, and the machine learning models used for classification used SVM and KNN. As a result, it was confirmed that SVM showed relatively higher accuracy than KNN and was more useful in diagnosing the presence of left ventricular diastolic dysfunction. In future research, it is expected that classification performance can be further improved by adding various indicators that evaluate not only TI but also cardiac function and securing more data. Furthermore, it is expected to be used as basic data to predict and classify other diseases and solve the problem of insufficient medical manpower compared to the increasing number of tests.

• Journal of Korean Society of Environmental Engineers
• /
• v.29 no.3
• /
• pp.348-356
• /
• 2007

Statistical and probabilistic method was used in the analysis of data, which is the most effective one in describing the various natures, and the methodology relating the results with the design was developed. Influents and effluents of three treatment plants were analyzed and the focus was made on BOD, COD, SS, IN, TP The fluctuations of influent such as BOD, COD, SS were extremely large and their standard deviations(st.dev) were more than 10 mg/L. but those of TN, TP were small; the st.dev was 6.6 mg/L for TN, 0.6 mg/L for TP, respectively. But, effluent concentration showed consistent pattern regardless of the influent fluctuations, the st.dev was ranged between 0.28 and 4.48 mg/L. Effluent distributional characteristics were as follows; BOD, COD were distributed normally, but SS, TN, and TP, log-normally; unsymmetric and skewed to the right. The coefficient of reliability(COR) based on the results of statistics of data was introduced to evaluate the process performance an4 to reflect the process performance to the process design. The coefficient of reliability relates the design value(the goal) with the standards and it can be used in operating treatment facilities under a certain reliability level and/or in evaluating the reliability of the treatment facilities on operation. Each treated water quality of effluent showed the half of water quality standards in the level of 50% percentile and all treatment plant was achieved 100% probability of water quality standards. It was concluded that the variability of the process performance should be reflected to the design procedure and the standards through the analysis based on the statistics and the probability.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.24 no.1
• /
• pp.36-42
• /
• 2018

This paper considers the Marine Traffic Risk Assessment for fixed and moving targets, which threaten officers during a voyage. The Collision Risk Assessment Formula was calculated based on a dynamic ship domain considering the length, speed and maneuvering capability of a vessel. In particular, the Navigation Risk Assessment Model that is used to quantitatively index the effect of a ship's size, speed, etc. has been reviewed and improved using a hybrid combination of a vessel's dynamic area and the Collision Risk Assessment Formula. Accordingly, a new type of Marine Traffic Risk Assessment Model has been suggested giving consideration to the Speed Length Ratio, which was not sufficiently reflected in the existing Risk Assessment Model. The larger the Speed Length Ratio (dimensionless speed), the higher the CJ value. That is, the CJ value is presented well by the Speed Length Ratio. When the Speed Length Ratio is large, states ranging from [Caution], [Warning], [Dangerous] or [Very Dangerous] are presented from a greater distance than when the Speed Length Ratio is small. The results of this study, can be used for route and port development, including dangerous route avoidance, optimum route planning, breakwater width, bridge span, etc. as well as the development of costal navigation safety charts. This research is also applicable for the selection of optimum ship routing and the prevention of collisions for smart ships such as autonomous vessels.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_3
• /
• pp.1475-1490
• /
• 2021

Detecting oil spill area using statistical characteristics of SAR images has limitations in that classification algorithm is complicated and is greatly affected by outliers. To overcome these limitations, studies using neural networks to classify oil spills are recently investigated. However, the studies to evaluate whether the performance of model shows a consistent detection performance for various oil spill cases were insufficient. Therefore, in this study, two CNNs (Convolutional Neural Networks) with basic structures(Simple CNN and U-net) were used to discover whether there is a difference in detection performance according to the structure of CNN and distribution characteristics of oil spill. As a result, through the method proposed in this study, the Simple CNN with contracting path only detected oil spill with an F1 score of 86.24% and U-net, which has both contracting and expansive path showed an F1 score of 91.44%. Both models successfully detected oil spills, but detection performance of the U-net was higher than Simple CNN. Additionally, in order to compare the accuracy of models according to various oil spill cases, the cases were classified into four different categories according to the spatial distribution characteristics of the oil spill (presence of land near the oil spill area) and the clarity of border between oil and seawater. The Simple CNN had F1 score values of 85.71%, 87.43%, 86.50%, and 85.86% for each category, showing the maximum difference of 1.71%. In the case of U-net, the values for each category were 89.77%, 92.27%, 92.59%, and 92.66%, with the maximum difference of 2.90%. Such results indicate that neither model showed significant differences in detection performance by the characteristics of oil spill distribution. However, the difference in detection tendency was caused by the difference in the model structure and the oil spill distribution characteristics. In all four oil spill categories, the Simple CNN showed a tendency to overestimate the oil spill area and the U-net showed a tendency to underestimate it. These tendencies were emphasized when the border between oil and seawater was unclear.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.31 no.6
• /
• pp.493-500
• /
• 2019

In this study, sensitivity analysis of sediment transport scaling factors in Delft3D-Morphology was performed to examine the effect those parameters on simulation results of cross-shore profile changes. For numerical experiments, one-year wave time series data which were observed in 2018 on the Maengbang coast in Gangwon prefecture were applied as external force. Bathymetric data observed in January and October of the same year were used as initial bathymetric data and annual bathymetric change data, respectively. The simulation performance of the model was evaluated based on the Brier Skill Score index for each part by dividing an arbitrary cross section within the calculation domain into the onshore and offshore parts. As a result, it was found thet the f_BED variable has a slight effect on the simulation results. The f_BEDW and f_SUSW variables show good simulation performance in onshore part when the value less than 0.5 is applied and vice versa. Among the experimental conditions, the optimal combinations of variables are f_BED = 1.0, f_BEDW = 1.0, f_SUSW = 0.1 for the onshore region and f_BED = 1.0, f_BEDW = 1.0, f_SUSW = 0.5 for the offshore region. However, since these combinations were derived based on the observation data on Maengbang beach in 2018, users should be careful when applying those results to other areas.