• Journal of Korean Tunnelling and Underground Space Association
• /
• v.25 no.6
• /
• pp.555-567
• /
• 2023

The cracks in the tunnel are currently determined through visual inspections conducted by inspectors based on images acquired using tunnel imaging acquisition systems. This labor-intensive approach, relying on inspectors, has inherent limitations as it is subject to their subjective judgments. Recently research efforts have actively explored the use of deep learning to automatically detect tunnel cracks. However, most studies utilize public datasets or lack sufficient objectivity in the analysis process, making it challenging to apply them effectively in practical operations. In this study, we selected test datasets consisting of images in the same format as those obtained from the actual inspection system to perform an objective evaluation of deep learning models. Additionally, we introduced ensemble techniques to complement the strengths and weaknesses of the deep learning models, thereby improving the accuracy of crack detection. As a result, we achieved high recall rates of 80%, 88%, and 89% for cracks with sizes of 0.2 mm, 0.3 mm, and 0.5 mm, respectively, in the test images. In addition, the crack detection result of deep learning included numerous cracks that the inspector could not find. if cracks are detected with sufficient accuracy in a more objective evaluation by selecting images from other tunnels that were not used in this study, it is judged that deep learning will be able to be introduced to facility safety inspection.

• The Journal of Sustainable Design and Educational Environment Research
• /
• v.23 no.1
• /
• pp.1-11
• /
• 2024

Complex school facilities are being considered to meet increased public demands for culture and welfare in Korea, given the decreasing population. In this context, Gyeongi-do Hwaseong City's E-umteo is recognized as a relatively well-operated school complex. Therefore, this study considered seven E-umteo branches as case studies to examine the operations of educational programs and understand the techniques employed in the spatial configuration of E-umteo. To this end, field observations and interviews with facility operators were conducted. The case analysis results revealed that educational programs could be classified into three types: learning sharing , community communication, and lifelong learning. Furthermore, the learning sharing type was classified into education and physical education while the community communication type was classified into the community and convenience types. Meanwhile, lifelong learning was identified as the most actively used type by differentiating specialized programs. With regard to the spatial composition between the school and the "pitcher," only the connection and independent types were observed, and no integral type was discovered. Therefore, integrated future studies are mandated.

• Korean Journal of Remote Sensing
• /
• v.40 no.2
• /
• pp.203-217
• /
• 2024

The Land Use, Land-Use Change and Forestry (LULUCF) sector of the National Greenhouse Gas Inventory is crucial for obtaining data on carbon sinks, necessitating accurate estimations. This study analyzes cases of countries applying the LULUCF sector at the Tier 3 level to propose enhanced methodologies for carbon sink estimation. In nations like Japan and Western Europe, satellite spatial information such as SPOT, Landsat, and Light Detection and Ranging (LiDAR)is used alongside national statistical data to estimate LULUCF. However, in Korea, the lack of land use change data and the absence of integrated management by category, measurement is predominantly conducted at the Tier 1 level, except for certain forest areas. In this study, Space-borne LiDAR Global Ecosystem Dynamics Investigation (GEDI) was used to calculate forest canopy heights based on Relative Height 100 (RH100) in the cities of Icheon, Gwangju, and Yeoju in Gyeonggi Province, Korea. These canopy heights were compared with the 1:5,000 scale forest maps used for the National Inventory Report in Korea. The GEDI data showed a maximum canopy height of 29.44 meters (m) in Gwangju, contrasting with the forest type maps that reported heights up to 34 m in Gwangju and parts of Icheon, and a minimum of 2 m in Icheon. Additionally, this study utilized Ordinary Least Squares(OLS)regression analysis to compare GEDI RH100 data with forest stand heights at the eup-myeon-dong level using ArcGIS, revealing Standard Deviations (SDs)ranging from -1.4 to 2.5, indicating significant regional variability. Areas where forest stand heights were higher than GEDI measurements showed greater variability, whereas locations with lower tree heights from forest type maps demonstrated lower SDs. The discrepancies between GEDI and actual measurements suggest the potential for improving height estimations through the application of high-resolution remote sensing techniques. To enhance future assessments of forest biomass and carbon storage at the Tier 3 level, high-resolution, reliable data are essential. These findings underscore the urgent need for integrating high-resolution, spatially explicit LiDAR data to enhance the accuracy of carbon sink calculations in Korea.

• Korean Journal of Environment and Ecology
• /
• v.38 no.1
• /
• pp.67-78
• /
• 2024

The term ecosystem services refers to natural ecosystems' benefits to humans. Various models have been developed and applied to quantify ecosystem services. Habitat quality assessment is a widely used leading InVEST ecosystem service model. In Korea, habitat quality assessment is conducted for national parks. For habitat quality assessment, the initial value of habitat quality must be used to assess the sensitivity to threats, which varies depending on the country and application field. Therefore, an expert survey (AHP) was conducted based on previous habitat quality assessments in national parks to adjust the sensitivity, the initial value for the habit quality assessment. As a result of the AHP, 18 items were adjusted, including 10 items, such as natural grassland and unarranged fields, upward and 8 items, such as rivers and ponds, downward. Based on the adjusted sensitivity results, the habitat quality of Bukhansan National Park and Gyeryongsan National Park (urban type), Gyeongju National Park (historic type), Hallyeohaesang National Park (ocean type), and Jirisan National Park and Seoraksan National Park (mountain type) were adjusted. The results of the analysis showed that the habitat quality of urban dry areas and water bodies distributed in the national parks was reflected in the habitat quality assessment. In the future, it will be possible to evaluate the habitat quality of natural parks using this standard.

• Information Systems Review
• /
• v.22 no.4
• /
• pp.21-39
• /
• 2020

Recently, deep learning has shown high performance in various applications such as pattern analysis and image classification. Especially known as a difficult task in the field of machine learning research, stock market forecasting is an area where the effectiveness of deep learning techniques is being verified by many researchers. This study proposed a deep learning Convolutional Neural Network (CNN) model to predict the direction of stock prices. We then used the feature selection method to improve the performance of the model. We compared the performance of machine learning classifiers against CNN. The classifiers used in this study are as follows: Logistic Regression, Decision Tree, Neural Network, Support Vector Machine, Adaboost, Bagging, and Random Forest. The results of this study confirmed that the CNN showed higher performancecompared with other classifiers in the case of feature selection. The results show that the CNN model effectively predicted the stock price direction by analyzing the embedded values of the financial data

• Journal of the Korean GEO-environmental Society
• /
• v.25 no.2
• /
• pp.5-14
• /
• 2024

Reinforced concrete bridge decks are the first to be damaged by vehicle loads and rain infiltration. Concrete deterioration primarily occurs owing to the corrosion of rebars and other metal components by chlorides used for snow and ice melting. The structural condition and concrete deterioration of the bridge decks within the pavement were evaluated using ground-penetrating radar (GPR) survey data. To evaluate concrete deterioration in bridges, it is necessary to develop GPR data analysis techniques to accurately identify deteriorated locations and rebar positions. GPR exploration involves the acquisition of reflection and diffraction wave signals due to differences in radar wave propagation velocity in geotechnical media. Therefore, a full-waveform inversion (FWI) method was developed to evaluate the deterioration of reinforced concrete bridge decks by estimating the radar wave propagation velocity in geotechnical media using GPR data. Numerical experiments using a GPR velocity model confirmed the deterioration phenomena of bridge decks, such as concrete delamination and rebar corrosion, verifying the applicability of the developed technology. Moreover, using the synthetic GPR data, FWI facilitates the determination of rebar positions and concrete deterioration locations using inverted velocity images.

• Korean Journal of Clinical Laboratory Science
• /
• v.55 no.4
• /
• pp.244-252
• /
• 2023

The timeliness and accuracy of test results are crucial factors for clinicians to decide and promptly administer effective and targeted antimicrobial therapy, especially in life-threatening infections or when vital organs and functions, such as sight, are at risk. Further research is needed to refine and optimize matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS)-based assays to obtain accurate and reliable results in the shortest time possible. MALDI-TOF MS-based bacterial identification focuses primarily on techniques for isolating and purifying pathogens from clinical samples, the expansion of spectral libraries, and the upgrading of software. As technology advances, many MALDI-based microbial identification databases and systems have been licensed and put into clinical use. Nevertheless, it is still necessary to develop MALDI-TOF MS-based antimicrobial-resistance analysis for comprehensive clinical microbiology characterization. The important applications of MALDI-TOF MS in clinical research include specific application categories, common analytes, main methods, limitations, and solutions. In order to utilize clinical microbiology laboratories, it is essential to secure expertise through education and training of clinical laboratory scientists, and database construction and experience must be maximized. In the future, MALDI-TOF mass spectrometry is expected to be applied in various fields through the use of more powerful databases.

• Journal of the Korea Institute of Building Construction
• /
• v.24 no.1
• /
• pp.157-168
• /
• 2024

The frequent occurrence of structural failures at building construction sites in Korea has underscored the critical role of rigorous oversight in the inspection and management of construction projects. As mandated by prevailing regulations and standards, onsite supervision by designated supervisors encompasses thorough documentation of construction quality, material standards, and the history of any reconstructions, among other factors. These reports, predominantly consisting of unstructured data, constitute approximately 80% of the data amassed at construction sites and serve as a comprehensive repository of quality-related information. This research introduces the SL-QPA model, which employs text mining techniques to preprocess supervision reports and establish a sentiment dictionary, thereby enabling the quantification of quality performance. The study's findings, demonstrating a statistically significant Pearson correlation between the quality performance scores derived from the SL-QPA model and various legally defined indicators, were substantiated through a one-way analysis of variance of the correlation coefficients. The SL-QPA model, as developed in this study, offers a supplementary approach to evaluating the quality performance of building construction projects. It holds the promise of enhancing quality inspection and management practices by harnessing the wealth of unstructured data generated throughout the lifecycle of construction projects.

• Journal of the Korean Society of Radiology
• /
• v.81 no.3
• /
• pp.620-631
• /
• 2020

Purpose We aimed to assess local tumor progression (LTP) rate and associated prognostic factors in 92 patients who underwent radiofrequency ablation (RFA) using saline-perfused electrodes to treat hepatocellular carcinoma (HCC) (≤ 5 cm). Materials and Methods Total 92 patients with 148 HCCs were treated with RFA using saline-perfused electrodes, from 2009 to 2015. We retrospectively evaluated technical success, technique efficacy, and LTP rates. Potential prognostic factors for LTP were perivascular tumor, subphrenic tumor, artificial ascites, tumor size (≥ 2 cm), and previous treatment of transarterial chemoembolization. Analysis was performed by lesion, rather than by person. Results During follow-up period from 1 to 97.4 months, total cumulative LTP rates were 7.9%, 11.4%, and 14.6% at 1, 3, and 5 years, respectively. These values were significantly higher in the perivascular (35.1%; p = 0.009) and subphrenic group (38.9%; p = 0.002) at 5-year. We did not observe any significant difference in LTP according to other prognostic factors (p > 0.05). Conclusion RFA with saline-perfused electrode is a safe and effective treatment modality for HCC (≤ 5 cm), with lower LTP rates. Nevertheless, perivascular and subphrenic HCCs demonstrated higher LTP rate than other sites. It is imperative to note that perivascular and subphrenic location of HCC are associated with a high risk of local recurrence, despite the use of saline-perfused electrodes.

• Information Systems Review
• /
• v.18 no.4
• /
• pp.89-106
• /
• 2016

The latest movement in live video streaming service is mobile personal broadcasting (MPB), which refers to consumers accessing the service through social media with mobile devices, such as smartphones and tablet PCs. This service is possible through the advancements in mobile video technology and platforms. Features such as enhanced user interaction, personalization, and real-time broadcasting, combined with a greater variety of content, have led to the development of MPB. The increase in MPB users calls for research, including that on the hedonic motivational angle. This study aims to assess MPB users' intrinsic motives through the hedonic-motivation system adoption model (HMSAM) using seven factors: joy, temporal dissociation, escapism, focused immersion, perceived ease of use, perceived usefulness and intention to watch. Survey data collected from 154 samples were analyzed with statistical techniques, such as structural equation modeling. Results showed that time dissociation, escapism, and perceived ease of use have a positive relationship with heightened enjoyment. Joy significantly affects focused immersion and intention to watch. Escapism also had a statistically significant influence on focused immersion. This study contributes to the advancement of the MPB study under the HMSAM theoretical framework and offers practical suggestions to managers to enhance MPB content viewership.