• Journal of the Korean Society of Radiology
• /
• v.14 no.7
• /
• pp.991-1001
• /
• 2020

Supervised deep learning technologies for improving the image quality of computed tomography (CT) need a lot of training data. When input images have different characteristics with training images, the technologies cause structural distortion in output images. In this study, an imaging model based on the deep reinforcement learning (DRL) was developed for overcoming the drawbacks of the supervised deep learning technologies and reducing noise in CT images. The DRL model was consisted of shared, value and policy networks, and the networks included convolutional layers, rectified linear unit (ReLU), dilation factors and gate rotation unit (GRU) in order to extract noise features from CT images and improve the performance of the DRL model. Also, the quality of the CT images obtained by using the DRL model was compared to that obtained by using the supervised deep learning model. The results showed that the image accuracy for the DRL model was higher than that for the supervised deep learning model, and the image noise for the DRL model was smaller than that for the supervised deep learning model. Also, the DRL model reduced the noise of the CT images, which had different characteristics with training images. Therefore, the DRL model is able to reduce image noise as well as maintain the structural information of CT images.

• Journal of the Korean Geotechnical Society
• /
• v.39 no.12
• /
• pp.23-35
• /
• 2023

The Ministry of Agriculture, Food and Rural Affairs has announced design standards for disaster-resilient greenhouses capable of resisting wind speeds with a 30-year frequency to respond to the destruction of greenhouses caused by strong winds. However, many greenhouses are still being maintained or newly installed as conventional standard facilities for the supply type. In these supply-type greenhouses, a small pile called a steel peg is used as reinforcement to resist wind-induced damage. The wind resistance of steel pegs varies depending on the soil environment and installation method. In this study, a correlation analysis was performed between the wind resistance of steel pegs installed in loam and sandy loam, using a soil hardness meter. To estimate the pull-out force of steel pegs based on soil water content and compaction, soil compaction tests and laboratory soil box and field tests were performed. The soil compaction degree was measured using a soil hardness meter that could easily confirm soil compaction. This was used to analyze the correlation between the soil compaction degree in the tests. In addition, a correlation analysis was performed between the pull-out force of steel pegs in the soil box and field. The findings of this study will be useful in predicting the pull-out force of steel pegs based on the method of steel peg installation and environmental changes.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.22 no.2
• /
• pp.5-25
• /
• 2011

This study aims to identify the most effective statistical feature selecting method and context window size for word sense disambiguation using supervised methods. In this study, features were selected by four different methods: information gain, document frequency, chi-square, and relevancy. The result of weight comparison showed that identifying the most appropriate features could improve word sense disambiguation performance. Information gain was the highest. SVM classifier was not affected by feature selection and showed better performance in a larger feature set and context size. Naive Bayes classifier was the best performance on 10 percent of feature set size. kNN classifier on under 10 percent of feature set size. When feature selection methods are applied to word sense disambiguation, combinations of a small set of features and larger context window size, or a large set of features and small context windows size can make best performance improvements.

• Journal of Broadcast Engineering
• /
• v.25 no.1
• /
• pp.48-57
• /
• 2020

Recently, with the rapid development of image recognition technology, the demand for object analysis in road CCTV videos is increasing. In this paper, we propose a method that can adaptively find the counting line for traffic analysis in road CCTV videos. First, vehicles on the road are detected, and the corresponding positions of the detected vehicles are modeled as the two-dimensional pointwise Gaussian map. The paths of vehicles are estimated by accumulating pointwise Gaussian maps on successive video frames. Then, we apply clustering and linear regression to the accumulated Gaussian map to find the principal direction of the road, which is highly relevant to the counting line. Experimental results show that the proposed method for detecting the counting line is effective in various situations.

• Journal of KIISE:Software and Applications
• /
• v.34 no.6
• /
• pp.519-529
• /
• 2007

In this paper, we investigate the use of fuzzy rules for efficient intrusion detection. We use evolutionary algorithm to optimize the set of fuzzy rules for intrusion detection by constructing fuzzy decision trees. For efficient execution of evolutionary algorithm we use supervised clustering to generate an initial set of membership functions for fuzzy rules. In our method both performance and complexity of fuzzy rules (or fuzzy decision trees) are taken into account in fitness evaluation. We also use evaluation with data partition, membership degree caching and zero-pruning to reduce time for construction and evaluation of fuzzy decision trees. For performance evaluation, we experimented with our method over the intrusion detection data of KDD'99 Cup, and confirmed that our method outperformed the existing methods. Compared with the KDD'99 Cup winner, the accuracy was increased by 1.54% while the cost was reduced by 20.8%.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.33-45
• /
• 2011

In this paper, we present a method for extracting video objects efficiently by using the modified graph cut algorithm based on contour information. First, we extract objects at the first frame by an automatic object extraction algorithm or the user interaction. To estimate the objects' contours at the current frame, motion information of objects' contour in the previous frame is analyzed. Block-based histogram back-projection is conducted along the estimated contour point. Each color model of objects and background can be generated from back-projection images. The probabilities of links between neighboring pixels are decided by the logarithmic based distance transform map obtained from the estimated contour image. Energy of the graph is defined by predefined color models and logarithmic distance transform map. Finally, the object is extracted by minimizing the energy. Experimental results of various test images show that our algorithm works more accurately than other methods.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.4
• /
• pp.993-1000
• /
• 2016

Semi-supervised learning makes it easy to use an unlabeled data in the supervised learning such as classification. Applying the semi-supervised learning on the regression analysis, we propose two methods for a better regression function estimation. The proposed methods have been assumed different marginal densities of independent variables and different smoothing parameters in unlabeled and labeled data. We shows that the overfitted pilot estimator should be used to achieve the fastest convergence rate and unlabeled data may help to improve the convergence rate with well estimated smoothing parameters. We also find the conditions of smoothing parameters to achieve optimal convergence rate.

• Journal of the Korean Geographical Society
• /
• v.44 no.3
• /
• pp.395-409
• /
• 2009

The objective of this paper is to illustrate that kriging can provide an effective framework both for integrating remote sensing data and for uncertainty modeling through a case study of sediment grain size mapping with remote sensing data. Landsat TM data which show reasonable relationships with grain size values are used as secondary information for sediment grain size mapping near the eastern part of Anmyeondo and Cheonsuman bay. The case study results showed that uncertainty attached to prediction at unsampled locations was significantly reduced by integrating remote sensing data through the analysis of conditional variance from conditional cumulative distribution functions. It is expected that the kriging-based approach presented in this paper would be efficient integration and analysis methodologies for any environmental thematic mapping using secondary information as well as sediment grain size mapping.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_1
• /
• pp.1187-1198
• /
• 2021

Sea wind isrecently drawing attraction as one of the sources of renewable energy. Thisstudy describes a new method to produce a 10 m resolution sea wind field using Sentinel-1 images and low-resolution NWP (Numerical Weather Prediction) data with artificial intelligence technique. The experiment for the South East coast in Korea, 2015-2020,showed a 40% decreased MAE (Mean Absolute Error) than the generic CMOD (C-band Model) function, and the CC (correlation coefficient) of our method was 0.901 and 0.826, respectively, for the U and V wind components. We created 10m resolution sea wind maps for the study area, which showed a typical trend of wind distribution and a spatially detailed wind pattern as well. The proposed method can be applied to surveying for wind power and information service for coastal disaster prevention and leisure activities.

• Journal of Advanced Navigation Technology
• /
• v.25 no.1
• /
• pp.8-16
• /
• 2021

In this paper, locations of vertiports were selected and evaluated to operate urban air mobility (UAM) in the Seoul metropolitan area. Demand data were analyzed using the data from the survey of commuting population and were marked on a map using MATLAB. To cluster the data, the K-means algorithm function built in MATLAB was used to identify the center of the cluster to as the location of vertiports, and using the silhouette technique, the accuracy and reliability of the clustering were evaluated. The locations of the selected vertiports were also identified using satellite maps to ensure that the locations of the selected vertiports were suitable for the actual vertiport location, and, if the location was not appropriate, final vertiports were selected through the repositioning process.