OPTIMIZATION OF CUBESAT-BASED HEAT MAPPING SYSTEM FOR EARLY DETECTION OF FOREST FIRE
Abstract
Wildfires are an increasingly frequent ecological and social disaster due to climate change and human activities. Major fires like the one that recently hit Los Angeles demonstrate the importance of rapid and accurate early detection to save lives, mitigate environmental damage, and protect infrastructure. Existing conventional satellite imaging systems still suffer from limitations in terms of spatial resolution, recapture duration, and high operational costs. In this study, a high-resolution thermal imaging-based CubeSat system was developed specifically for local heat mapping and real-time monitoring of wildfires. Furthermore, five detection models were tested using public datasets to identify the best performing detection model. This research includes the design of real-time data collection using Cubesats equipped with the best-performing models for detecting forest fires and heat anomalies to support disaster response, sustainable resource management, and environmental conservation. This innovation not only improves the accuracy of early detection of forest fires and heat anomalies but also enables efficient monitoring of geological resources. By supporting rapid decision-making in disaster mitigation and sustainable resource management, this system is an environmentally-oriented solution that strengthens resilience to future ecological crises.
References
Giglio, L., Schroeder, W., & Justice, C. O. (2016). The Collection 6 MODIS active fire detection algorithm and fire products. Remote Sensing of Environment, 178, 31–41.
Justice, C. O., Townshend, J. R. G., Holben, B. N., & Tucker, C. J. (2002). Analysis of the phenology of global vegetation using MODIS data. Remote Sensing of Environment, 83(1–2), 198–208.
Kramer, H., Rossi, C., & Müller, R. (2018). Thermal imaging for wildfire detection using CubeSat platforms. International Journal of Remote Sensing, 39(15–16), 5500–5515.
Kulu, E. (2014). Small satellite constellations for Earth observation: Opportunities and challenges. Acta Astronautica, 102(1), 128–136.
Li, X., Zhang, Y., & Chen, W. (2019). Integration of multi-source satellite data for early wildfire detection. Remote Sensing, 11(14), 1654.
Lotfi, A., & Amini, M. (2021). BAT algorithm optimization for hyperparameter tuning in machine learning models. Expert Systems with Applications, 165, 113906.
Moustris, K. P., Doulamis, A. D., & Doulamis, N. D. (2016). Remote sensing technologies for forest fire detection and management. Computers and Electronics in Agriculture, 123, 10–22.
Nowak, W., & Stepinski, T. F. (2018). Deep learning for forest fire detection using multispectral satellite imagery. ISPRS Journal of Photogrammetry and Remote Sensing, 142, 121–133.
Swartwout, M. (2013). The first one hundred CubeSats: A statistical look. Journal of Small Satellites, 2(2), 213–233.
Woellert, K., Ehrenfreund, P., Ricco, A., & Hertzfeld, H. (2011). Cubesats: Cost-effective science and technology platforms for emerging and developing nations. Advances in Space Research, 47(4), 663–684.
Xu, H., Liu, Y., & Li, Z. (2020). Early forest fire detection based on machine learning and remote sensing data. Forests, 11(6), 635.
Zhang, H., Wang, L., & Li, Y. (2019). Fire detection in forests using XGBoost and satellite thermal imagery. Remote Sensing Letters, 10(12), 1201–1210.
Zhang, Y., Li, X., & Wang, L. (2020). AI-assisted fire detection in forests using satellite imagery. IEEE Access, 8, 56789–56799.
Zhou, B., & Song, S. (2021). Optimization of XGBoost hyperparameters using metaheuristic algorithms for environmental applications. Environmental Modelling & Software, 139, 104991.
.jpg){kind=logo}
