Sanika. S. Jagtap,
Swapnil. S. Salekar,
Tanvi. S. Ravalekar,
Akshay. R. Ghule,
- Student, Department of Computer Engineering, Rajgad Technical Campus, Pune, Maharashtra, India
- Student, Department of Computer Engineering, Rajgad Technical Campus, Pune, Maharashtra, India
- Student, Department of Computer Engineering, Rajgad Technical Campus, Pune, Maharashtra, India
- Student, Department of Computer Engineering, Rajgad Technical Campus, Pune, Maharashtra, India
Abstract
To create a precise forecast of the wildfire, it is crucial to possess the capability to recognize it and anticipate how it will distribute. The devastation of vegetation, the loss of assets, the rise in greenhouse gases, the extinction of Numerous animal species and even human fatalities can all result from wildfires. To trim back this danger, there must be a mechanism capable of detecting a fire the moment it begins to escalate so that we can extinguish the fires before they escalate into a wildfire. Generally, conventional methods depend on machine learning. models to predict the presence of a fire. We must determine how the fire is spreading if we want to protect. additional lives and inventory, and this isn’t proving very useful. Consequently, to achieve this, the proposed model categorizes fires in images sourced from NASA’s database utilizing the given latitude and longitude utilizing a Fire pixel. After obtaining the satellite image, the YOLO model is used to detect the flame. Finally, we can provide the wildfire’s development path following the implementation of the decision-making model. In addition to detection, the proposed system incorporates a fire spread prediction module that estimates the future progression of wildfire using environmental parameters such as humidity, wind speed. Furthermore, an automated email alert mechanism is generated to notify concerned authorities in real time by sending fire detection results along with image attachments. This combines approach enhances situational awareness and supports faster decision making in emergency scenarios.
Keywords: Fire pixel model, YOLO model, decision making, email alert, wildfire detection
[This article belongs to International Journal on Drones ]
References
- Chen X, et al. Wildland fire detection and monitoring using drone-collected RGB/IR image dataset. IEEE Access. 2022. doi:10.1109/ACCESS.2022.322285.
- Singh VK, Singh C, Raza H. Event classification and intensity discrimination for forest fire inference with IoT. IEEE Sens J. 2022;22(9):8869-8880. doi:10.1109/JSEN.2022.3163155.
- Kizilkaya B, Ever E, Yatbaz HY, Yazici A. An effective framework for detecting forest fires using heterogeneous wireless multimedia sensor networks. ACM Trans Multimed Comput Commun Appl. 2022;18(2):47.
- Dampage U, Bandaranayake L, Wanasinghe R, Kottahachchi K, Jayasanka B. Forest fire detection system using wireless sensor networks and machine learning. Sci Rep. 2022;12:46. doi:10.1038/s41598-021-03882-9.
- Tran DQ, Park M, Jeon Y, Bak J, Park S. Forest fire response system using deep learning with weather data and CCTV images. IEEE Access. 2022;10:66061-66071. doi:10.1109/ACCESS.2022.3184707.
- ChrispinJiji A, Prasad KDV, Jeyaraman N, Sahu DN, Begum AY, Patil NJ. IoT-based automatic forest fire detection using machine learning. Ann For Res. 2022.
- Lasaponara R, Fattore C, Modica G. Tortoli Ogliastra fire (Sardinia): a case study using Sentinel-1 and Sentinel-2. IEEE Geosci Remote Sens Lett. 2023. doi:10.1109/LGRS.2023.3324945.
- Rui L, Jiaqing Z, Yang H. Dehazing model of extra-high voltage converter station based on two-stage attention. IEEE Access. 2023;11:133246-133255. doi:10.1109/ACCESS.2023.3327438.
- Wang L, Zhang H, Zhang Y, Hu K. A deep learning-based experiment on forest wildfire detection in machine vision course. IEEE Access. 2023;11:32671-32681. doi:10.1109/ACCESS.2023.3262701.
International Journal on Drones
| Volume | 02 | |
| Issue | 02 | |
| Received | 07/04/2026 | |
| Accepted | 04/08/2026 | |
| Published | 20/08/2026 | |
| Publication Time | 135 Days |