• Journal of the Korea institute for structural maintenance and inspection
• /
• v.24 no.6
• /
• pp.77-84
• /
• 2020

In this paper, we developed a damage evaluation technique that can determine the damage location of a long-sized structure such as a cable-stayed bridge, and verified the performance of the developed technique through experiments. The damage assessment method aims to extract data that can evaluate the damage of the structure without the undamage data and can determine the damage location only by analyzing the response data of the structure. To complete this goal, we developed a damage assessment technique that considers variability based on the IMD theory, which is a statistical pattern recognition technique, to identify the damage location. To complete this goal, we developed a damage assessment technique that considers variability based on the IMD theory, which is a statistical pattern recognition technique, to identify the damage location. To evaluate the performance of the developed technique experimentally, cable damage experiments were conducted on model cable-stayed bridges. As a result, the damage assessment method considering variability automatically outputs the damageless data according to external force, and it is confirmed that the performance of extracting information that can determine the damage location of the cable through the analysis of the outputted damageless data and the measured damage data is shown.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.1
• /
• pp.29-40
• /
• 2021

Risk factors that threaten public safety such as crime, fire, and traffic accidents have spatial characteristics. Since each region has different dangerous environments, it is necessary to analyze the spatial pattern of risk factors for each sector such as traffic accident, fire, crime, and living safety. The purpose of this study is to analyze the spatial distribution pattern of local safety level index, which act as an index that rates the safety level of each sector (traffic accident, fire, crime, living safety, suicide, and infectious disease) for basic local governments across the nation. The following analysis tools were used to analyze the spatial autocorrelation of local safety level index : Global Moran's I, Local Moran's I, and Getis-Ord's G⁎i. The result of the analysis shows that the distribution of safety level on traffic accidents, fire, and suicide tends to be more clustered spatially compared to the safety level on crime, living safety, and infectious disease. As a result of analyzing significant spatial correlations between different regions, it was found that the Seoul metropolitan areas are relatively safe compared to other cities based on the integrated index of local safety. In addition, hot spot analysis using statistical values from Getis-Ord's G⁎i derived three hot spots(Samchuck, Cheongsong-gun, and Gimje) in which safety-vulnerable areas are clustered and 15 cold spots which are clusters of areas with high safety levels. These research findings can be used as basic data when the government is making policies to improve the safety level by identifying the spatial distribution and the spatial pattern in areas with vulnerable safety levels.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.23 no.1
• /
• pp.82-88
• /
• 2021

Sun-induced fluorescence (SIF) retrieval using remote sensing technique has been used in an effort to understand the photosynthetic efficiency and stress condition of vegetation. Although optical devices and SIF retrieval methodologies were established in order to retrieve SIF, the SIF measurements are domestically sparse. SIF data of paddy rice w as measured in Naju, South Korea from June 10, 2020 to October 5, 2020. The SIFs based red (O2A) and far-red (O2B) w ere retrieved using a spectral fitting method and an improved Fraunhofer line depth, and photosynthetically active radiation was also produced. In addition, the SIF data was filtered considering solar zenith angle, saturation conditions, the rapid and sudden change of solar irradiance, and sun glint. The provided SIF data can help to understand a SIF product and the filtering method of SIF data can contribute to producing high-quality SIF data.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_1
• /
• pp.1367-1377
• /
• 2020

The purpose of this study is to assess the seismic vulnerability of buildings in Gyeongju city starting with the earthquake that occurred in the city on September 12, 2016, and produce a seismic vulnerability map. 11 influence factors related to geotechnical, physical, and structural indicators were selected to assess the seismic vulnerability, and these were applied as independent variables. For a dependent variable, location data of the buildings that were actually damaged in the 9.12 Gyeongju Earthquake was used. The assessment model was constructed based on random forest (RF) as a mechanic study method and support vector machine (SVM), and the training and test dataset were randomly selected with a ratio of 70:30. For accuracy verification, the receiver operating characteristic (ROC) curve was used to select an optimum model, and the accuracy of each model appeared to be 1.000 for RF and 0.998 for SVM, respectively. In addition, the prediction accuracy was shown as 0.947 and 0.926 for RF and SVM, respectively. The prediction values of the entire buildings in Gyeongju were derived on the basis of the RF model, and these were graded and used to produce the seismic vulnerability map. As a result of reviewing the distribution of building classes as an administrative unit, Hwangnam, Wolseong, Seondo, and Naenam turned out to be highly vulnerable regions, and Yangbuk, Gangdong, Yangnam, and Gampo turned out to be relatively safer regions.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.4
• /
• pp.125-132
• /
• 2021

As the layers of artificial neural network deepens, and the dimension of data used as an input increases, there is a problem of high arithmetic operation requiring a lot of arithmetic operation at a high speed in the learning and recognition of the neural network (NN). Thus, this study proposes a data dimensionality reduction method to reduce the dimension of the input data in the NN. The proposed Line-segment Feature Analysis (LFA) algorithm applies a gradient-based edge detection algorithm using median filters to analyze the line-segment features of the objects existing in an image. Concerning the extracted edge image, the eigenvalues corresponding to eight kinds of line-segment are calculated, using 3×3 or 5×5-sized detection filters consisting of the coefficient values, including [0, 1, 2, 4, 8, 16, 32, 64, and 128]. Two one-dimensional 256-sized data are produced, accumulating the same response values from the eigenvalue calculated with each detection filter, and the two data elements are added up. Two LFA256 data are merged to produce 512-sized LAF512 data. For the performance evaluation of the proposed LFA algorithm to reduce the data dimension for the recognition of handwritten numbers, as a result of a comparative experiment, using the PCA technique and AlexNet model, LFA256 and LFA512 showed a recognition performance respectively of 98.7% and 99%.

• Journal of the Korean Geotechnical Society
• /
• v.37 no.5
• /
• pp.47-63
• /
• 2021

Recently, the demand for three-dimensional (3D) underground maps from the perspective of digital twins and the demand for linkage utilization are increasing. However, the vastness of national geotechnical survey data and the uncertainty in applying geostatistical techniques pose challenges in modeling underground regional geotechnical characteristics. In this study, an optimal learning model based on multi-layer perceptron (MLP) was constructed for 3D subsurface lithological and geotechnical classification in Seoul, South Korea. First, the geotechnical layer and 3D spatial coordinates of each borehole dataset in the Seoul area were constructed as a geotechnical database according to a standardized format, and data pre-processing such as correction and normalization of missing values for machine learning was performed. An optimal fitting model was designed through hyperparameter optimization of the MLP model and model performance evaluation, such as precision and accuracy tests. Then, a 3D grid network locally assigning geotechnical layer classification was constructed by applying an MLP-based bet-fitting model for each unit lattice. The constructed 3D geotechnical layer map was evaluated by comparing the results of a geostatistical interpolation technique and the topsoil properties of the geological map.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.6
• /
• pp.243-250
• /
• 2021

With the increased development of 3D VR applications augmented by recent VR/AR/MR technologies and by the advance of 3D devices, interchangeability and portability of 3D data have become essential. 3D files should be processed in a standard data format for common usage between applications. Providing standardized libraries and data structures along with the standard file format means that a more efficient system organization is possible and unnecessary processing due to the usage of different file formats and data structures depending on the applications can be omitted. In order to provide the function of using a common data file and data structure, this research is intended to provide a programming binding tool for generating and storing standardized data so that various services can be developed by accessing the common 3D files. To achieve this, this paper defines a common data structure including classes and functions to access X3D files with a standardized scheme using the Python programming language. It describes the implementation of a Python language binding viewer, which is an X3D VR viewer for rendering standard X3D data files based on the language binding interface. The VR viewer includes Python based 3D scene libraries and a data structure for creation, modification, exchange, and transfer of X3D objects. In addition, the viewer displays X3D objects and processes events using the libraries and data structure.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.133-140
• /
• 2022

We investigate the performance of deep learning-based Korean language models on a task of predicting the score range of Korean essays written by foreign students. We construct a data set containing a total of 304 essays, which include essays discussing the criteria for choosing a job ('job'), conditions of a happy life ('happ'), relationship between money and happiness ('econ'), and definition of success ('succ'). These essays were labeled according to four letter grades (A, B, C, and D), and a total of eleven essay score range prediction experiments were conducted (i.e., five for predicting the score range of 'job' essays, five for predicting the score range of 'happiness' essays, and one for predicting the score range of mixed topic essays). Three deep learning-based Korean language models, KoBERT, KcBERT, and KR-BERT, were fine-tuned using various training data. Moreover, two traditional probabilistic machine learning classifiers, naive Bayes and logistic regression, were also evaluated. Experiment results show that deep learning-based Korean language models performed better than the two traditional classifiers, with KR-BERT performing the best with 55.83% overall average prediction accuracy. A close second was KcBERT (55.77%) followed by KoBERT (54.91%). The performances of naive Bayes and logistic regression classifiers were 52.52% and 50.28% respectively. Due to the scarcity of training data and the imbalance in class distribution, the overall prediction performance was not high for all classifiers. Moreover, the classifiers' vocabulary did not explicitly capture the error features that were helpful in correctly grading the Korean essay. By overcoming these two limitations, we expect the score range prediction performance to improve.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.141-148
• /
• 2022

Recently, reinforcement learning combined with deep neural network technology has achieved remarkable success in various fields such as board games such as Go and chess, computer games such as Atari and StartCraft, and robot object manipulation tasks. However, such deep reinforcement learning describes states, actions, and policies in vector representation. Therefore, the existing deep reinforcement learning has some limitations in generality and interpretability of the learned policy, and it is difficult to effectively incorporate domain knowledge into policy learning. On the other hand, dNL-RRL, a new relational reinforcement learning framework proposed to solve these problems, uses a kind of vector representation for sensor input data and lower-level motion control as in the existing deep reinforcement learning. However, for states, actions, and learned policies, It uses a relational representation with logic predicates and rules. In this paper, we present dNL-RRL-based policy learning for transportation mobile robots in a manufacturing environment. In particular, this study proposes a effective method to utilize the prior domain knowledge of human experts to improve the efficiency of relational reinforcement learning. Through various experiments, we demonstrate the performance improvement of the relational reinforcement learning by using domain knowledge as proposed in this paper.

• Knowledge Management Research
• /
• v.23 no.1
• /
• pp.47-67
• /
• 2022

Computer and Internet information technologies(ICTs) have changed the modern service industry and people's life style. In particular, the global spread of COVID-19 has attracted more attention to contact service types such as self-service technology. With the increase in labor costs and the enhancement of consumer self-awareness, more and more companies transfer part of their work to customers through their own service technology. This study seeks to answer the following questions. (1) Do technology difficulty and safety perception affect customer value recognition in the self-service technologies? (2) Does customer value recognition influence the intention to use such technologies continuously? This study conducted an empirical analysis with 327 samples to validate the influence of self-service characteristics(technology difficulty and safety perception) on customer value recognition and continuous utilization intentions. Also, it analyzes the moderating effects of age and frequency of use on the relationship between self-service characteristics and customer value recognition. The study results show that the technology difficulty does not affect the customer's perceived value recognition; and the higher the customer's value recognition, the higher the intention of continuous use.