RESEARCH
📚 Representative Works
* Staying the mouse over the screen stops the movement
Fire Risk Assessment
Wildfires can cause extensive damage, including deforestation and environmental pollution, even from a single event. Restoring damaged areas requires significant effort and time. Thus, assessing wildfire risk in advance is crucial for proper preparedness and response. Current wildfire risk indices are mostly based on weather variables related to wildfire ignition, and analyses of various environmental factors related to wildfire risk remain insufficient.
Kang et al. (2019) and (2020) addressed this gap by developing an hourly forest fire risk index (HFRI) with 1 km resolution for South Korea, incorporating accessibility, fuel, time, and weather factors through the Catboost machine learning algorithm. HFRI, created from an ensemble model combining integrated and meteorological models, provided more spatially detailed and seasonally stable simulations compared to earlier indices. Feature contributions, analyzed using Shapley Additive Explanations (SHAP), identified relative humidity, elevation, road density, and population density as key factors, highlighting the importance of accessibility in human-caused wildfire risks.
Kang et al., 2019
Kang et al., 2020
📝 Publications
Last updated : 2023/03/12
Kang, Y., Jang, E., Im, J., & Kwon, C. (2022). A deep learning model using geostationary satellite data for forest fire detection with reduced detection latency. GIScience & Remote Sensing, 59(1), 2019-2035.
Lee, J., Kim, W., Im, J., Kwon, C., & Kim, S. (2021). Detection of Forest Fire Damage from Sentinel-1 SAR Data through the Synergistic Use of Principal Component Analysis and K-means Clustering. Korean Journal of Remote Sensing, 37(5), 1373-1387.
Sim, S., Kim, W., Lee, J., Kang, Y., Im, J., Kwon, C., & Kim. S. (2020). Wildfire Severity Mapping Using Sentinel Satellite Data Based on Machine Learning Approaches. Korean Journal of Remote Sensing, 36(5), 1109-1123.
Kang, Y., Jang, E., Im, J., Kwon, C., & Kim, S. (2020). Developing a New Hourly Forest Fire Risk Index Based on Catboost in South Korea. Applied Sciences, 10(22), 8213.
Kang, Y., Park, S., Jang, E., Im, J., Kwon, C., & Lee, S. (2019). Spatio-temporal enhancement of forest fire risk index using weather forecast and satellite data in South Korea. Journal of the Korean Association of Geographic Information Studies, 22, 116–130.
Park, S., Son, B., Im, J., Lee, J., Lee, B., & Kwon, C. (2019). Development of Satellite-Based Drought Indices for Assessing Wildfire Risk. Korean Journal of Remote Sensing, 35(6), 1285-1298.
Park, H., Im, J., & Kim, M. (2019). Improvement of satellite-based estimation of gross primary production through optimization of meteorological parameters and high resolution land cover information at regional scale over East Asia. Agricultural and Forest Meteorology, 271, 180-192.
Jang, E., Kang, Y., Im, J., Lee, D. W., Yoon, J., & Kim, S. K. (2019). Detection and monitoring of forest fires using Himawari-8 geostationary satellite data in South Korea. Remote Sensing, 11(3), 271.
Lee, J., Im, J., Kim, K., & Quackenbush, L. J. (2018). Machine learning approaches for estimating forest stand height using plot-based observations and airborne LiDAR data. Forests, 9(5), 268.
Fang, F., Im, J., Lee, J., & Kim, K. (2016). An improved tree crown delineation method based on live crown ratios from airborne LiDAR data. GIScience & Remote Sensing, 53(3), 402-419.
Li, M., Im, J., Quackenbush, L. J., & Liu, T. (2014). Forest biomass and carbon stock quantification using airborne LiDAR data: A case study over Huntington Wildlife Forest in the Adirondack Park. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 7(7), 3143-3156.