• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.237-250
• /
• 2022

Structural health monitoring (SHM) plays an important role in ensuring the safety and functionality of critical civil infrastructure. In recent years, numerous researchers have conducted studies to develop computer vision and machine learning techniques for SHM purposes, offering the potential to reduce the laborious nature and improve the effectiveness of field inspections. However, high-quality vision data from various types of damaged structures is relatively difficult to obtain, because of the rare occurrence of damaged structures. The lack of data is particularly acute for fatigue crack in steel bridge girder. As a result, the lack of data for training purposes is one of the main issues that hinders wider application of these powerful techniques for SHM. To address this problem, the use of synthetic data is proposed in this article to augment real-world datasets used for training neural networks that can identify fatigue cracks in steel structures. First, random textures representing the surface of steel structures with fatigue cracks are created and mapped onto a 3D graphics model. Subsequently, this model is used to generate synthetic images for various lighting conditions and camera angles. A fully convolutional network is then trained for two cases: (1) using only real-word data, and (2) using both synthetic and real-word data. By employing synthetic data augmentation in the training process, the crack identification performance of the neural network for the test dataset is seen to improve from 35% to 40% and 49% to 62% for intersection over union (IoU) and precision, respectively, demonstrating the efficacy of the proposed approach.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.2
• /
• pp.210-217
• /
• 2022

Passive source depth estimation has been studied for decades since the source depth can be used for target classification, target tracking, etc. The purpose of this paper is to solve the problem of ambiguity in the previous paper [S.-il. Cho et al. (in Korean), J. Acoust. Soc. Kr. 38, 120-127 (2019)] that source depth is estimated in two points. The patterns of phase shift of Channel Impulse Response(CIR) reflected in ocean surface and bottom is used for removing ambiguity of the source depth estimation, and after removing ambiguity, source depth is estimated at one point through the intersection of CIR. In order to extract CIR in case of unknown source signal and continuous signal or noise, Ray-based blind deconvolution is used. The proposed algorithm is demonstrated through numerical simulation in ocean waveguide.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_3
• /
• pp.949-965
• /
• 2023

In this study, we propose a method to monitor the surface area of agricultural reservoirs in South Korea using Sentinel-1 synthetic aperture radar images and the deep learning model, Swin Transformer. Utilizing the Google Earth Engine platform, datasets from 2017 to 2021 were constructed for seven agricultural reservoirs, categorized into 700 K-ton, 900 K-ton, and 1.5 M-ton capacities. For four of the reservoirs, a total of 1,283 images were used for model training through shuffling and 5-fold cross-validation techniques. Upon evaluation, the Swin Transformer Large model, configured with a window size of 12, demonstrated superior semantic segmentation performance, showing an average accuracy of 99.54% and a mean intersection over union (mIoU) of 95.15% for all folds. When the best-performing model was applied to the datasets of the remaining three reservoirsfor validation, it achieved an accuracy of over 99% and mIoU of over 94% for all reservoirs. These results indicate that the Swin Transformer model can effectively monitor the surface area of agricultural reservoirs in South Korea.

• Tunnel and Underground Space
• /
• v.11 no.3
• /
• pp.257-269
• /
• 2001

Numerical simulation is performed to understand the structural behavior of an underground radwaste repository, assumed to be located at the depth of 500 m, in a granitic rock mats, in which a fault intersects the roof of the repository cavern. Two dimensional universal distinct element code, UDEC is used in the analysis. The numerical model includes a granitic rock mass, a canister with PWR spent fuels surrounded by the compacted bentonite inside the deposition hole, and the mixed bentonite backfilled in the rest of the space within the repository cavern. The structural behavior of three different cases, each case with a fault of an angle of $33^{\circ},\;45^{\circ},\;and\;58^{\circ}$ passing through the cavern roof-wall intersection, has been compared. And then fro the case with the $45^{\circ}$ fault, the hydro-mechanical, thermo-mechanical, and thermo-hydro-mechanical interaction behavior have been studied. The effect of the time-dependent decaying heat, from the radioactive materials in PWR spent fuels, on the repository and its surroundings has been studied. The groundwater table is assumed to be located 10m below the ground surface, and a steady state flow algorithm is used.

• The Journal of Engineering Geology
• /
• v.4 no.2
• /
• pp.153-168
• /
• 1994

The status of groundwater development in Chungnam was studied with geological characteristics according to the measured data of Korean Rural Development Corporation. The data of 212 survey wells were used for the relation between catchment area and water discharge, and the data of 344 development wells for the relationships between well depth and discharge, between casing depth and discharge, between rock type and discharge, and the relation with lineaments density. The relationship between the catchment area and discharge does not show any special trend, and it is understood that groundwater of hard rock mass is not so much influenced by the surface catchment area. The relationship between well depth and discharge shows two different trends; discharge increasing with depth for alluvial groundwater, but no certain trend between depth and discharge for groundwater of hard rock zone. Discharge increases linearly with the casing depth, and it is reliable because the casing was installed in the weathered zone against well destruction. Generally the rock type does not show any difference of discharge, but the crystalline rocks such as granite and gneiss yield a little more discharge than the more porous rocks such as sedimentary rock or schist. It suggests that the effect of fracture zone is a major governing factor. In Hongsong and Puyo, there are similar in rock type and casing depth, but the big difference in average discharge. The big discharge of Hongsong is concordant with the higher intersection density and longer length of lineament in Hongsong than those of Puyo. Therefore the groundwater development strategy should be focused on the micro topography analysis and geophysical survey for the understanding of the fracture zone rather than catchment area or rock type.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.18 no.1
• /
• pp.170-181
• /
• 2015

A research on extracting croplands located nearby coastal areas using the spatial information data sets is the important task for managing the agricultural products in coastal areas. This research aims to extract the various croplands(croplands on mountains and croplands on plain areas) located nearby coastal areas using the KOMPSAT-2 imagery, the high-resolution satellite imagery, and the airborne topographic LiDAR(Light Detection And Ranging) data acquired in coastal areas of Uljin, Korea. Firstly, the NDVI(Normalized Difference Vegetation Index) imagery is generated from the KOMPSAT-2 imagery, and the vegetation areas are extracted from the NDVI imagery by using the appropriate threshold. Then, the DSM(Digital Surface Model) and DEM(Digital Elevation Model) are generated from the LiDAR data by using interpolation method, and the CHM(Canopy Height Model) is generated using the differences of the pixel values of the DSM and DEM. Then the plain areas are extracted from the CHM by using the appropriate threshold. The low slope areas are also extracted from the slope map generated using the pixel values of the DEM. Finally, the areas of intersection of the vegetation areas, the plain areas and the low slope areas are extracted with the areas higher than the threshold and they are defined as the croplands located nearby coastal areas. The statistical results show that 85% of the croplands on plain areas and 15% of the croplands on mountains located nearby coastal areas are extracted by using the proposed methodology.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.4
• /
• pp.173-181
• /
• 2007

In this paper, we propose a moving target localization algorithm using acoustic spectrograms. A time-versus-frequency spectrogram provide a information of trajectory of the moving target in underwater. For a source at sufficiently long range from a receiver, broadband striation patterns seen in spectrogram represents the mutual interference between modes which reflected by surface and bottom. The slope of the maximum intensity striation is influenced by waveguide invariant parameter ${\beta}$ and distance between target and sensor. When more than two sensors are applied to measure the moving ship-radited noise, the slope and frequency of the maximum intensity striation are depend on distance between target and receiver. We assumed two sensors to fixed point then form a circle of apollonios which set of all points whose distances from two fixed points are in a constant ratio. In case of three sensors are applied, two circle form an intersection point so coordinates of this point can be estimated as a position of target. To evaluates a performance of the proposed localization algorithm, simulation is performed using acoustic propagation program.

• The Journal of the Petrological Society of Korea
• /
• v.25 no.3
• /
• pp.195-209
• /
• 2016

The understanding of geometric complexity of strike-slip Fault system can be an important factor to control fault reactivation and surface rupture propagation under the regional stress regime. The Kumamoto earthquake was caused by dextral reactivation of the Futagawa-Hinagu Fault system under the E-W maximum horizontal principal stress. The earthquakes are a set of earthquakes, including a foreshock earthquake with a magnitude 6.2 at the northern tip of the Hinagu Fault on April 14, 2016 and a magnitude 7.0 mainshock which generated at the intersection of the two faults on April 16, 2016. The hypocenters of the main shock and aftershocks have moved toward NE direction along the Futagawa Fault and terminated at Mt. Aso area. The intersection of the two faults has a similar configuration of ${\lambda}$-fault. The geometries and kinematics, of these faults were comparable to the Yansan-Ulsan Fault system in SE Korea. But slip rate is little different. The results of age dating show that the Quaternary faults distributed along the northern segment of the Yangsan Fault and the Ulsan Fault are younger than those along the southern segment of the Yansan Fault. This result is well consistent with the previous study with Column stress model. Thus, the seismic activity along the middle and northern segment of the Yangsan Fault and the Ulsan Fault might be relatively active compared with that of the southern segment of the Yangsan Fault. Therefore, more detailed seismic hazard and paleoseismic studies should be carried out in this area.

• Korean Journal of Remote Sensing
• /
• v.39 no.2
• /
• pp.193-205
• /
• 2023

A large amount of floating debris from land-based sources during heavy rainfall has negative social, economic, and environmental impacts, but there is a lack of monitoring systems for floating debris accumulation areas and amounts. With the recent development of artificial intelligence technology, there is a need to quickly and efficiently study large areas of water systems using drone imagery and deep learning-based object detection models. In this study, we acquired various images as well as drone images and trained with You Only Look Once (YOLO)v5s and the recently developed YOLO7 and YOLOv8s to compare the performance of each model to propose an efficient detection technique for land-based floating debris. The qualitative performance evaluation of each model showed that all three models are good at detecting floating debris under normal circumstances, but the YOLOv8s model missed or duplicated objects when the image was overexposed or the water surface was highly reflective of sunlight. The quantitative performance evaluation showed that YOLOv7 had the best performance with a mean Average Precision (intersection over union, IoU 0.5) of 0.940, which was better than YOLOv5s (0.922) and YOLOv8s (0.922). As a result of generating distortion in the color and high-frequency components to compare the performance of models according to data quality, the performance degradation of the YOLOv8s model was the most obvious, and the YOLOv7 model showed the lowest performance degradation. This study confirms that the YOLOv7 model is more robust than the YOLOv5s and YOLOv8s models in detecting land-based floating debris. The deep learning-based floating debris detection technique proposed in this study can identify the spatial distribution of floating debris by category, which can contribute to the planning of future cleanup work.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1413-1425
• /
• 2023

The increasing frequency of wildfires due to climate change is causing extreme loss of life and property. They cause loss of vegetation and affect ecosystem changes depending on their intensity and occurrence. Ecosystem changes, in turn, affect wildfire occurrence, causing secondary damage. Thus, accurate estimation of the areas affected by wildfires is fundamental. Satellite remote sensing is used for forest fire detection because it can rapidly acquire topographic and meteorological information about the affected area after forest fires. In addition, deep learning algorithms such as convolutional neural networks (CNN) and transformer models show high performance for more accurate monitoring of fire-burnt regions. To date, the application of deep learning models has been limited, and there is a scarcity of reports providing quantitative performance evaluations for practical field utilization. Hence, this study emphasizes a comparative analysis, exploring performance enhancements achieved through both model selection and data design. This study examined deep learning models for detecting wildfire-damaged areas using Landsat 8 satellite images in California. Also, we conducted a comprehensive comparison and analysis of the detection performance of multiple models, such as U-Net and High-Resolution Network-Object Contextual Representation (HRNet-OCR). Wildfire-related spectral indices such as normalized difference vegetation index (NDVI) and normalized burn ratio (NBR) were used as input channels for the deep learning models to reflect the degree of vegetation cover and surface moisture content. As a result, the mean intersection over union (mIoU) was 0.831 for U-Net and 0.848 for HRNet-OCR, showing high segmentation performance. The inclusion of spectral indices alongside the base wavelength bands resulted in increased metric values for all combinations, affirming that the augmentation of input data with spectral indices contributes to the refinement of pixels. This study can be applied to other satellite images to build a recovery strategy for fire-burnt areas.