• Journal of the Korea institute for structural maintenance and inspection
• /
• v.23 no.3
• /
• pp.83-91
• /
• 2019

The purpose of this study is to assess a fire-damaged concrete structure using a digital camera and image processing software. To simulate it, mortar and paste samples of W/C=0.5(general strength) and 0.3(high strength) were put into an electric furnace and simulated from $100^{\circ}C$ to $1000^{\circ}C$. Here, the paste was processed into a powder to measure CIELAB chromaticity, and the samples were taken with a digital camera. The RGB chromaticity was measured by color intensity analyzer software. As a result, the residual compressive strength of W/C=0.5 and 0.3 was 87.2 % and 86.7 % at the heating temperature of $400^{\circ}C$. However there was a sudden decrease in strength at the temperature above $500^{\circ}C$, while the residual compressive strength of W/C=0.5 and 0.3 was 55.2 % and 51.9 % of residual strength. At the temperature $700^{\circ}C$ or higher, W/C=0.5 and W/C=0.3 show 26.3% and 27.8% of residual strength, so that the durability of the structure could not be secured. The results of $L^*a^*b$ color analysis show that $b^*$ increases rapidly after $700^{\circ}C$. It is analyzed that the intensity of yellow becomes strong after $700^{\circ}C$. Further, the RGB analysis found that the histogram kurtosis and frequency of Red and Green increases after $700^{\circ}C$. It is analyzed that number of Red and Green pixels are increased. Therefore, it is deemed possible to estimate the degree of damage by checking the change in yellow($b^*$ or R+G) when analyzing the chromaticity of the fire-damaged concrete structures.

• Journal of Korea Water Resources Association
• /
• v.41 no.6
• /
• pp.565-574
• /
• 2008

Fire-enhanced soil hydrophobicity often increases runoff and erosion in the mountain hillslope following severe wildfires. Estimation techniques for WEPP's parameters were studied in burnt mountain slopes. In burnt mountain slopes, the model over-predicted runoff in the small runoff and under-predicted runoff in the great runoff, and in the lower sediment runoff it had a tendency to over-predict soil loss. The effective hydraulic conductivity was most sensitive in the WEPP's runoff and its sediment runoff was mainly effected by the effective hydraulic conductivity, initial saturation, rill erodibility, and interrill erodibility. To improve the applicability of the WEPP, the adjustment coefficient of effective hydraulic conductivity was defined for runoff and the adjustment coefficient of rill erodibility and interrill erodibility was presented for sediment runoff. The adjustment coefficient of effective hydraulic conductivity in wildfire mountain slopes increased with maximum rainfall intensity of single storm and the vegetation height index. The adjustment coefficients of rill erodibility depended on soil components of size distribution curve and total rainfall depths in single storm. The adjustment coefficients of interrill erodibility decreased with increases of maximum rainfall intensity and vegetation height index. These results may be used in the application of WEPP model for wildfire mountain slopes.

• Korean Journal of Remote Sensing
• /
• v.37 no.6_1
• /
• pp.1731-1738
• /
• 2021

The objective of this study was to extract individual trees and tree heights using UAV drone images. The study site was Gongju national university experiment forest, located in Yesan-gun, Chungcheongnam-do. The thinning intensity study sites consisted of 40% thinning, 20% thinning, 10% thinning and control. The image was filmed by using the "Mavic Pro 2" model of DJI company, and the altitude of the photo shoot was set at 80% of the overlay between 180m pictures. In order to prevent image distortion, a ground reference point was installed and the end lap and side lap were set to 80%. Tree heights were extracted using Digital Surface Model (DSM) and Digital Terrain Model (DTM), and individual trees were split and extracted using object-based analysis. As a result of individual tree extraction, thinning 40% stands showed the highest extraction rate of 109.1%, while thinning 20% showed 87.1%, thinning 10% showed 63.5%, and control sites showed 56.0% of accuracy. As a result of tree height extraction, thinning 40% showed 1.43m error compared with field survey data, while thinning 20% showed 1.73 m, thinning 10% showed 1.88 m, and control sites showed the largest error of 2.22 m.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_3
• /
• pp.979-995
• /
• 2023

As wildfires are difficult to predict, real-time monitoring is crucial for a timely response. Geostationary satellite images are very useful for active fire detection because they can monitor a vast area with high temporal resolution (e.g., 2 min). Existing satellite-based active fire detection algorithms detect thermal outliers using threshold values based on the statistical analysis of brightness temperature. However, the difficulty in establishing suitable thresholds for such threshold-based methods hinders their ability to detect fires with low intensity and achieve generalized performance. In light of these challenges, machine learning has emerged as a potential-solution. Until now, relatively simple techniques such as random forest, Vanilla convolutional neural network (CNN), and U-net have been applied for active fire detection. Therefore, this study proposed an active fire detection algorithm using state-of-the-art (SOTA) deep learning techniques using data from the Advanced Himawari Imager and evaluated it over East Asia and Australia. The SOTA model was developed by applying EfficientNet and lion optimizer, and the results were compared with the model using the Vanilla CNN structure. EfficientNet outperformed CNN with F1-scores of 0.88 and 0.83 in East Asia and Australia, respectively. The performance was better after using weighted loss, equal sampling, and image augmentation techniques to fix data imbalance issues compared to before the techniques were used, resulting in F1-scores of 0.92 in East Asia and 0.84 in Australia. It is anticipated that timely responses facilitated by the SOTA deep learning-based approach for active fire detection will effectively mitigate the damage caused by wildfires.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.22 no.1
• /
• pp.33-45
• /
• 2019

Compound disaster is the type that increases the impact affected by two or more hazard events, and attention to compound disaster and multi-hazards risk is growing due to potential damages which are difficult to predict. The objective of this study is to analyze the possible impacts of post-fire landslide scenario quantitatively by using TRIGRS (Transient Rainfall Infiltration and Grid-Based Regional Slope-Stability Analysis), a physics-based landslide model. In the case of wildfire, soil organic material and density are altered, and saturated hydraulic conductivity decrease because of soil exposed to high temperature. We have included the change of soil saturated hydraulic conductivity into the TRIGRS model through literature review. For a case study, we selected the area of $8km^2$ in Pyeongchang County. The landslide modeling process was calibrated before simulate the post-wildfire impact based on landslide inventory data to reduce uncertainty. As a result, the mean of the total factor of safety values in the case of landslide was 2.641 when rainfall duration is 1 hour with rainfall intensity of 100mm per day, while the mean value for the case of post-wildfire landslide was lower to 2.579, showing potential landslide occurrence areas appear more quickly in the compound disaster scenario. This study can be used to prevent potential losses caused by the compound disaster such as post-wildfire debris flow or landslides.

• Journal of Industrial Convergence
• /
• v.21 no.7
• /
• pp.29-39
• /
• 2023

In the recent period, there have been numerous earthquakes both domestically and internationally, and buildings in South Korea are particularly vulnerable to seismic design and earthquake damage. Therefore, the objective of this study is to discover an effective method for assessing the seismic vulnerability of buildings and conducting a density analysis of high-risk structures. The aim is to model this approach and validate it using data from pilot area(Seoul). To achieve this, two modeling techniques were employed, of which the predictive accuracy of the statistical analysis technique was 87%. Among the machine learning techniques, Random Forest Model exhibited the highest predictive accuracy, and the accuracy of the model on the Test Set was determined to be 97.1%. As a result of the analysis, the district rating revealed that Gwangjin-gu and Songpa-gu were relatively at higher risk, and the density analysis of at-risk buildings predicted that Seocho-gu, Gwanak-gu, and Gangseo-gu were relatively at higher risk. Finally, the result of the statistical analysis technique was predicted as more dangerous than those of the machine learning technique. However, considering that about 18.9% of the buildings in Seoul are designed to withstand the Seismic intensity of 6.5 (MMI), which is the standard for seismic-resistant design in South Korea, the result of the machine learning technique was predicted to be more accurate. The current research is limited in that it only considers buildings without taking into account factors such as population density, police stations, and fire stations. Considering these limitations in future studies would lead to more comprehensive and valuable research.

• 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.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.11
• /
• pp.104-109
• /
• 2017

The flooding and deterioration of water quality caused by urbanization and climate change are becoming more serious. In order to respond to this, studies on low impact development (LID) technology, which is designed to restore the hydrological system of the urban basin to its natural state, have been actively pursued all over the world, The announcement of the low carbon green growth law, hydrophilic area special law, etc., highlights the importance of technology such as the LID method. However, whereas various developments have been made in relation to the current LID element technology, there has been little research designed to verify its effectiveness. In this study, we analyzed the optimum spatial distribution of pitcher fire pitcher packing in parking lots using the K - LIDM model to verify the effectiveness of the low impact development (LID) method in the early stages. Using the eight package scenario and the three rain intensity scenarios, it was found that the lower 40% pitcher packaging results in an approximately 90% spill reduction effect, as in the case of the whole pitcher's package. The confirmation of these analyses and experimental verification is expected to ensure that the actual pitcher packaging will be used as a basis for arranging LID facilities such as urban planning and housing development in the future.