Edge Computing-Based Real-Time Forest Fire Detection Using UAV Thermal and Color Images
Bibliographic record
Abstract
Fire detection using aerial platform is an important technology for forest surveillance. But the real-time detection capability is still a challenging problem. In this article, an edge computing-based real-time forest fire detection strategy is designed using the uncrewed aerial vehicle (UAV). The objective is to improve the timely response capability and the detection accuracy for early stage small fires. The thermal and color images obtained from the onboard cameras are registered to the same scale and merged with appropriate proportions. These preprocessed dual-modal images become the input for training the fire detection network model. To deploy this model on the resource-constrained UAV edge computing device, it is compressed and accelerated to reduce size and enhance efficiency. Experiments based on self-made UAV dual-modal images of simulated fire scenarios and public datasets derived from real forest environments are conducted to validate the accuracy and speed of the proposed method. Experimental results show that, on the self-made dataset, the mAP is 93.76%, and the inference speed reaches 34.6 FPS on the ground computer. On the public dataset, the mAP is 97.53%, and the inference speed reaches 16 FPS on the edge computing device iCrest 2-s. Compared to several state-of-the-art methods, our proposed method achieves a good tradeoff between accuracy and speed.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".