• Tunnel and Underground Space
• /
• v.31 no.5
• /
• pp.374-384
• /
• 2021

In this study, we propose a new approach for automatic fracture detection in CT scan images of rock specimens. This approach is built on top of two-stage object detection deep learning algorithm called Faster R-CNN with a major modification of using rotated bounding box. The use of rotated bounding box plays a key role in the future work to overcome several inherent difficulties of fracture segmentation relating to the heterogeneity of uninterested background (i.e., minerals) and the variation in size and shape of fracture. Comparing to the commonly used bounding box (i.e., axis-align bounding box), rotated bounding box shows a greater adaptability to fit with the elongated shape of fracture, such that minimizing the ratio of background within the bounding box. Besides, an additional benefit of rotated bounding box is that it can provide relative information on the orientation and length of fracture without the further segmentation and measurement step. To validate the applicability of the proposed approach, we train and test our approach with a number of CT image sets of fractured granite specimens with highly heterogeneous background and other rocks such as sandstone and shale. The result demonstrates that our approach can lead to the encouraging results on fracture detection with the mean average precision (mAP) up to 0.89 and also outperform the conventional approach in terms of background-to-object ratio within the bounding box.

• The Journal of Korean Society for Radiation Therapy
• /
• v.30 no.1_2
• /
• pp.169-176
• /
• 2018

Purpose : The purpose of this study is to evaluate clinical applicability of Co-60 ViewRay treatment plan to increase the skin dose in case of high skin dose is required such as Malignant Fungating Wound By measuring the presence / absence of Bolus application and skin dose by the treatment device and comparing it Materials and Methods : Nine inner measuring points of 2.5 cm lattice arrangement and all 13 measuring points including upper and lower left and right measuring points touching the chest and skin were marked. After CT was taken, each treatment plan was formulated through Eclipse and ViewRay-TPS, and a Fixed beam-IMRT treatment plan was formulated so that the left chest V2Gy=95 % is delivered. Before measurement QED detector was calibrated and the QED detector was positioned at the 13 measurement points displayed on Phantom and surface dose of each treatment planner was measured using 5 mm Bolus application using True-beam and View-ray before and after, measure three times and compare each before applying 5 mm Bolus. Results : The surface dose of the Co-60 ViewRay and the linear accelerator appeared at $76.8%{\pm}5.2%$ vs. $67.3{\pm}%7.5%$ and the surface dose after application of 5 mm Bolus was $87.6%{\pm}8.9%$ vs. $80.3%{\pm}10.2%$ It was measured at 10.2 % (p<0.001). Conclusion : As a result of the surface dose measurement of each treatment instrument, Co-60 ViewRay confirmed that the surface dose reached 95.6 % of 6 MV Linac with conventional 5 mm bolus, despite not using Bolus (p<0.001). Also, by utilizing magnetic resonance images for each treatment, it is possible to observe the change in the treatment site without the problem of exposure, it is easy to formulate an adaptive treatment plan and it is easy to secure the skin dose, so the size In the case of Malignant Fungating Wound patients who need fast skin changes and need high skin doses, Co-60 ViewRay is considered to be more useful than linear accelerators.

• Economic and Environmental Geology
• /
• v.54 no.1
• /
• pp.1-19
• /
• 2021

This study reports the Ar-Ar dating results for the volcanic rocks from small volcanoes(oreum) of the Hallasan Nature Reserve. According to the age of 40Ar/39Ar, the volcanic activity of the Hallasan Natural Reserve was started from about 192 ka ago. The basaltic trachyandesite and trachyte located in the Y valley near the Eorimok in the western part of the Hallasan Natural Reserve represent an age of about 191~192 ka, showing the oldest record of volcanic activity in the Hallasan Natural Reserve. In the Hallasan Natural Reserve, the small volcanoes older than 100 ka are Y Valley in Eorimok area (192±5 and 191±5 ka), Dongsu-Ak (184±19 ka), Mansedongsan (153±5 ka), Janggumok-Orum (135±6 ka), Eoseungsaengak (123±9 ka), Samgagbong (105±2 ka). And the small volcanoes younger than 100 ka are Witbangae-Oreum, Seongneol-Oreum, Muljangol, Yeongsil, Bori-Ak, Witsenueun-Oreum, Witsejokeun-Oreum, Heugbuleun-Oreum, Bangae-Oreum, Albangae-Oreum, Witsebuleun-Oreum, Baengnokdam, Nongo-Ak. According to the eruption of trachytes, the Hallasan Natural Reserve can be interpreted as having about 8 volcanic activities. Among them, 4 volcanic activities are related with the formation of trachyte dome, such as Wanggwanneung, Samgakbong, Yeongsil, and Baengnokdam, and 4 volcanic activities are related with flow or dyke of trachyte. The volcanic activity at the Hallasan Natural Reserve was started from northwest area, to in the southern area, and in the eastern area, and finally volcanic activity related to the formation of Baengnokdam.

• (The)Study of the Eastern Classic
• /
• no.36
• /
• pp.163-202
• /
• 2009

The poems of Shin Suk-ju can be divided into two periods around the political change by King Sejo; his poems from the former period show that the poet enjoyed abundant pride and leisurely spirit in the self-satisfying world driven by his determination to maintain a pure heart and save and awaken the people during the reign of King Sejong. His ideology of awakening the people, however, was the product of his heroic consciousness to achieve immortal fame. It was his heroic consciousness and determination to sacrifice his life for fame according to the mandate from Heaven that made him join the political change by King Sejo. His poems from the latter period clearly reveal that the way of his life to pursue fame didn't bring him spiritual satisfaction and happiness. He confessed that his conscience was destroyed as he joined King Sejo in his political change and the deeds he achieved and further his life itself were all in vain. He lost the values or orders he pursued, which caused instability in his life. Facing such instability head-on, he argued that right and wrong, true and false, and good and evil mentioned in the world were all subjective and groundless. Furthermore, he realized all the things and creatures of the world were nothing but phantoms. Those perceptions he had were based on Madhyamaka of Buddhism. Going through such a thinking process, the poet wrote about his mentality of a false reputation with ideal mentality. Heroic consciousness, Buddhist thinking, and pursuit of mentality of a false reputation found in his poems make also frequent appearances in the poems by major literary men in the latter half of the 15th century such as Seo Geo-jeong. His serious searches to overcome his conscientious agony and sense of futility about life had influences on the attitudes toward life and literature of the official literary men of the times. Seong Hyeon's statement that the major literary figures of the times inherited the literary tradition of Shin Suk-ju was not a rhetoric by courtesy.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.6
• /
• pp.265-274
• /
• 2021

Outlier detection research in ocean data has traditionally been performed using statistical and distance-based machine learning algorithms. Recently, AI-based methods have received a lot of attention and so-called supervised learning methods that require classification information for data are mainly used. This supervised learning method requires a lot of time and costs because classification information (label) must be manually designated for all data required for learning. In this study, an autoencoder based on unsupervised learning was applied as an outlier detection to overcome this problem. For the experiment, two experiments were designed: one is univariate learning, in which only SST data was used among the observation data of Deokjeok Island and the other is multivariate learning, in which SST, air temperature, wind direction, wind speed, air pressure, and humidity were used. Period of data is 25 years from 1996 to 2020, and a pre-processing considering the characteristics of ocean data was applied to the data. An outlier detection of actual SST data was tried with a learned univariate and multivariate autoencoder. We tried to detect outliers in real SST data using trained univariate and multivariate autoencoders. To compare model performance, various outlier detection methods were applied to synthetic data with artificially inserted errors. As a result of quantitatively evaluating the performance of these methods, the multivariate/univariate accuracy was about 96%/91%, respectively, indicating that the multivariate autoencoder had better outlier detection performance. Outlier detection using an unsupervised learning-based autoencoder is expected to be used in various ways in that it can reduce subjective classification errors and cost and time required for data labeling.

• Economic and Environmental Geology
• /
• v.55 no.5
• /
• pp.511-519
• /
• 2022

In order to develop an air-stripping based remediation process to remove the TOC (Total Organic Carbon) in groundwater around the underground LPG storage cavern, the laboratory scale experiments at various conditions (change of air injection volume and temperature, the application of ultrasonic treatment, etc.) for two types of groundwater (initial TOC concentration of 608 mg/L and 153 mg/L, respectively). From results of experiment, as the air injection rate for stripping into groundwater increased from 2 L/min to 11 L/min and as the air-stripping time increased from 1 hour to 24 hour, the TOC removal efficiency of air-stripping increased. However, the TOC concentration of treated groundwater was higher than the discharge tolerance limit (100 mg/L) even after 24 hour stripping at the maximum air injection rate of 11 L/min. The main compounds of the TOC in groundwater were identified as methanol and propane and the long stripping time (more than 24 hour) was needed to separate the methanol from groundwater because of the affinity between water and methanol. At 20℃ and 4 L/min of air injection, the TOC removal efficiency increased to 59.1% after 24 hour air-stripping. When the temperature of groundwater increased to 30℃ and 40℃, the TOC removal efficiency increased up to 80.0% and 82.8%, suggesting that more than 24 hour air-stripping at 40℃ is needed to lower the TOC concentration to below 100 mg/L and the additional TOC removal process as well as the air-stripping is necessary. When the temperature increased to 60℃ and the ultrasonic treatment was conjugated with the air-stripping, the TOC removal efficiency increased to 87.8% within 5 hour stripping and the final TOC concentration (72.4 mg/L) was satisfied with the TOC discharge tolerance limit. The TOC removal efficiency for groundwater having low TOC concentration (153 mg/L) also showed similar removal efficiency of 89.7% (the final TOC concentration: 18.9 mg/L). Results in this study supported that the air-stripping conjugated with the ultrasonic treatment could remove successfully the TOC in groundwater around the underground LPG strorage cavern.

• Journal of the Korean Geotechnical Society
• /
• v.22 no.6
• /
• pp.71-82
• /
• 2006

One of the most popular pre-reinforcement methods of tunnel heading in cohesionless soils would be the fore-polling of grouted pipes, known as RPUM (reinforced protective umbrella method) or UAM (umbrella arch method). This technique allows safe excavation even in poor ground conditions by creating longitudinal arch parallel to the tunnel axis as the tunnel advances. Some previous studies on the reinforcing effects have been performed using numerical methods and/or laboratory-based small scale model tests. The complexity of boundary conditions imposes difficulties in representing the tunnelling procedure in laboratory tests and theoretical approaches. Full-scale study to identify reinforcing effects of the tunnel heading has rarely been carried out so far. In this study, a large scale model testing for a tunnel in granular soils was performed. Reinforcing patterns considered are four cases, Non-Reinforced, Crown-Reinforced, Crown & Face-Reinforced, and Face-Reinforced. The behavior of ground and pipes as reinforcing member were fully measured as the surcharge pressure applied. The influences of reinforcing pattern, pipe length, and face reinforcement were investigated in terms of stress and displacement. It is revealed that only the Face-Reinforced has decreased sufficiently both vertical settlement in tunnel heading and horizontal displacement on the face. Vertical stresses along the tunnel axis were concentrated in tunnel heading from the test results, so the heading should be reinforced before tunnel advancing. Most of maximum axial forces and bending moments for Crown-reinforced were measured at 0.75D from the face. Also it should be recommended that the minimum length of the pipe is more than l.0D for crown reinforcement.

• Korean Journal of Remote Sensing
• /
• v.39 no.3
• /
• pp.325-338
• /
• 2023

Korean fir (Abies koreana Wilson) is one of the most important environmental indicator tree species for assessing climate change impacts on coniferous forests in the Korean Peninsula. However, due to the nature of alpine and subalpine regions, it is difficult to conduct regular field surveys of Korean fir, which is mainly distributed in regions with altitudes greater than 1,000 m. Therefore, this study analyzed the vegetation change trend of Korean fir using regularly observed remote sensing data. Specifically, normalized difference vegetation index (NDVI) from Moderate Resolution Imaging Spectroradiometer (MODIS), land surface temperature (LST), and precipitation data from Global Precipitation Measurement (GPM) Integrated Multi-satellitE Retrievalsfor GPM from September 2003 to 2020 for Hallasan and Jirisan were used to analyze vegetation changes and their association with environmental variables. We identified a decrease in NDVI in 2020 compared to 2003 for both sites. Based on the NDVI difference maps, areas for healthy vegetation and high mortality of Korean fir were selected. Long-term NDVI time-series analysis demonstrated that both Hallasan and Jirisan had a decrease in NDVI at the high mortality areas (Hallasan: -0.46, Jirisan: -0.43). Furthermore, when analyzing the long-term fluctuations of Korean fir vegetation through the Hodrick-Prescott filter-applied NDVI, LST, and precipitation, the NDVI difference between the Korean fir healthy vegetation and high mortality sitesincreased with the increasing LST and decreasing precipitation in Hallasan. Thissuggests that the increase in LST and the decrease in precipitation contribute to the decline of Korean fir in Hallasan. In contrast, Jirisan confirmed a long-term trend of declining NDVI in the areas of Korean fir mortality but did not find a significant correlation between the changes in NDVI and environmental variables (LST and precipitation). Further analyses of environmental factors, such as soil moisture, insolation, and wind that have been identified to be related to Korean fir habitats in previous studies should be conducted. This study demonstrated the feasibility of using satellite data for long-term monitoring of Korean fir ecosystems and investigating their changes in conjunction with environmental conditions. Thisstudy provided the potential forsatellite-based monitoring to improve our understanding of the ecology of Korean fir.

• Korean Journal of Mineralogy and Petrology
• /
• v.36 no.2
• /
• pp.147-166
• /
• 2023

This study aimed to develop and apply guidelines for calculating greenhouse gas emissions to activate the contribution of the Korea Institute of Geoscience and Mineral Resources (KIGAM) for institutional-level research activities. In addition, we intended to improve awareness by identifying greenhouse gas emissions from KIGAM's basic research and development (R&D) activities in fiscal 2022. Herein, the research plan and budget contents of individual projects were analyzed, whilst the boundaries and scopes of greenhouse gas emissions were determined, with 22 cases being derived as either direct, indirect, or other sources of emissions. Subsequently, research activity emissions were calculated by emission source. The greenhouse gas emissions of KIGAM's 2022 basic project R&D activities were 2,041.506 tCO2eq, of which direct emissions were 793.235 tCO2eq (38.86%), indirect emissions comprised 305.647 tCO2eq (14.97%), whilst other emissions were 942.624 tCO2eq (46.18%). In particular, greenhouse gas emissions per 100 million won in the KIGAM's basic projects for fiscal 2022 (a total of 96.661 billion won) was calculated as 2.11 tCO2eq, whilst greenhouse gas emissions per participating researcher (was 4.800 tCO2eq. Such calculations should be carried out annually rather than once and accumulated for at least 5 years. Accordingly, it will be possible to standardize specific matters that influence emissions according to differences in research field characteristics and methods, thus guiding greenhouse gas emission reduction management in the future and evaluating the contributions of Environmental, Social and Governance (ESG) management to the environmental sector.

• Journal of Korean Society of Forest Science
• /
• v.112 no.4
• /
• pp.426-441
• /
• 2023

This study investigated the predictive accuracy of a model of landslide displacement in Jecheon-si, where a great number of landslides were triggered by heavy rain on both natural (non-clear-cut) and clear-cut slopes during August 2020. This was accomplished by applying three flow direction methods (single flow direction, SFD; multiple flow direction, MFD; infinite flow direction, IFD) and the degree of root cohesion to an infinite slope stability equation. The application assumed that the soil saturation and any changes in root cohesion occurred following the timber harvest (clear-cutting). In the study area, 830 landslide locations were identified via landslide inventory mapping from satellite images and 25 cm resolution aerial photographs. The results of the landslide modeling comparison showed the accuracy of the models that considered changes in the root cohesion following clear-cutting to be improved by 1.3% to 2.6% when compared with those not considered in the area under the receiver operating characteristics (AUROC) analysis. Furthermore, the accuracy of the models that used the MFD algorithm improved by up to 1.3% when compared with the models that used the other algorithms in the AUROC analysis. These results suggest that the discriminatory application of the root cohesion, which considers changes in the vegetation condition, and the selection of the flow direction method may influence the accuracy of landslide predictive modeling. In the future, the results of this study should be verified by examining the root cohesion and its dynamic changes according to the tree species using the field hydrological monitoring technique.