Camera Ready Submission Oral Presentation.pptx

ICEEICT 24 Conference MIST

1. PyroVision : A Deep Learning Based Model for Wildfire Detection in Satellite Imagery Shoukat Alam Sifat Computer Science and Engineering United International University Mirza Hasan Computer Science and Engineering United International University Santi Brata Nath (Joy) [Presenting] Computer Science and Engineering United International University Md. Ahashan Habib Computer Science and Engineering United International University 347

2. Outline Problem Statement Motivation Research Objectives Methodology Result and Analysis Conclusion 2 03-May-2024 Email: [email protected]

3. Problem Statement Critical challenge : Accurate wildfire detection in satellite imagery. Current methods struggle : High false positives, low spatial resolution. Need: Deep learning model for wildfire identification. 3 03-May-2024 Email: [email protected]

4. Motivation 1. Environmental preservation and wildlife protection. 2. Human safety enhancement through timely wildfire detection. 3. Deep learning methods for improved disaster management. 4 03-May-2024 Email: [email protected]

5. Research Objectives Developing a method using deep learning techniques. Creating a model, PyroVision , capable of accurately identifying wildfires in satellite images. Improving false positive and false negative. Achieving high precision and recall. 5 03-May-2024 Email: [email protected]

6. Methodology 6 03-May-2024 Email: [email protected]

7. Result & Analysis Result : 7 03-May-2024 Email: [email protected] Model Accuracy Precision Recall F1 score FP FN 2D CNN 88.40% 99.35% 76.07% 86.16% 10 483 VGG19 55.41% 0% 0% 0% 1747 MobileNetV2 83.57% 81.75% 83.34% 82.54% 325 291 PyroVision 95.51% 95.53% 94.80% 95.16% 88 103

8. Result & Analysis Analysis : Temperature, humidity and wind speed are the most significant predictors of wildfire occurrence. Future research could focus on integrating real-time data streams, such as weather forecasts and satellite imagery, to improve predictive accuracy. Limitations included reliance on satellite image quality and scalability issues. 8 03-May-2024 Email: [email protected]

9. Conclusion PyroVision : framework for wildfire prediction. Achieved 95.51% predictive accuracy. Challenges: Data acquisition, scalability. Commitment to ongoing refinement as multiple environment. Significance: Safeguarding communities and reduce the devastating effects of fire. 9 03-May-2024 Email: [email protected]

10. Reference [1]N. T. Toan , P. Thanh Cong, N. Q. Viet Hung, and J. Jo, “A deep learning approach for early wildfire detection from hyperspectral satellite images,” in 2019 7th International Conference on Robot Intelligence Technology and Applications ( RiTA ), 2019, pp. 38–45 [2] G. L. James, R. B. Ansaf , S. S. Al Samahi , R. D. Parker, J. M. Cutler,R . V. Gachette , and B. I. Ansaf , “An efficient wildfire detection system for ai -embedded applications using satellite imagery,” Fire, vol. 6, no. 4, p. 169, 2023 [3] M. Shahid , S.-F. Chen, Y.-L. Hsu, Y.-Y. Chen, Y.-L. Chen, and K.-L. Hua, “Forest fire segmentation via temporal transformer from aerial images,” Forests, vol. 14, no. 3, p. 563, 2023. [4] A. Namburu , P. Selvaraj , S. Mohan, S. Ragavanantham , and E. T.Eldin , “Forest fire identification in uav imagery using x- mobilenet ,”Electronics, vol. 12, no. 3, p. 733, 2023. [5] J. R. Marlon, P. J. Bartlein , D. G. Gavin, C. J. Long, R. S. Anderson, C. E. Briles , K. J. Brown, D. Colombaroli , D. J. Hallett, M. J. Power et al., “Long-term perspective on wildfires in the western usa ,” Proceedings of the National Academy of Sciences, vol. 109, no. 9, pp. E535–E543, 2012. 10 03-May-2024 Email: [email protected]

11. Reference [6] C. Stewart, “The australian ‘black saturday ’ bushfires of 2009,” www.britannica.com/explore/savingearth/ the-australian-black-saturday-bushfires-of-2009, accessed: [19 Jan, 2024]. [7] E. J. Bowd , S. C. Banks, C. L. Strong, and D. B. Lindenmayer , “ Longterm impacts of wildfire and logging on forest soils,” Nature Geoscience, vol. 12, no. 2, pp. 113–118, 2019. [8] R. Ghali , M. Jmal , W. Souidene Mseddi , and R. Attia , “Recent advances in fire detection and monitoring systems: A review,” in Proceedings of the 8th International Conference on Sciences of Electronics, Technologies of Information and Telecommunications (SETIT’18), Vol. 1. Springer, 2020, pp. 332–340. [9] A. Nazir , H. Mosleh , M. Takruri , A.-H. Jallad , and H. Alhebsi , “Early fire detection: a new indoor laboratory dataset and data distribution analysis,” Fire, vol. 5, no. 1, p. 11, 2022. [10] Y. Sun, L. Jiang, J. Pan, S. Sheng, and L. Hao , “A satellite imagery smoke detection framework based on the mahalanobis distance for early fire identification and positioning,” International Journal of Applied Earth Observation and Geoinformation , vol. 118, p. 103257, 2023 11 03-May-2024 Email: [email protected]