Forest fire flame and smoke detection from UAV-captured images using fire-specific color features and multi-color space local binary pattern
Bibliographic record
Abstract
In recent years, the frequency and severity of forest fire occurrence have increased, compelling the research communities to actively search for early forest fire detection and suppression methods. Remote sensing using computer vision techniques can provide early detection from a large field of view along with providing additional information such as location and severity of the fire. Over the last few years, the feasibility of forest fire detection by combining computer vision and aerial platforms such as manned and unmanned aerial vehicles, especially low cost and small-size unmanned aerial vehicles, have been experimented with and have shown promise by providing detection, geolocation, and fire characteristic information. This paper adds to the existing research by proposing a novel method of detecting forest fire using color and multi-color space local binary pattern of both flame and smoke signatures and a single artificial neural network. The training and evaluation images in this paper have been mostly obtained from aerial platforms with challenging circumstances such as minuscule flame pixels, varying illumination and range, complex backgrounds, occluded flame and smoke regions, and smoke blending into the background. The proposed method has achieved F1 scores of 0.84 for flame and 0.90 for smoke while maintaining a processing speed of 19 frames per second. It has outperformed support vector machine, random forest, Bayesian classifiers and YOLOv3, and has demonstrated the capability of detecting challenging flame and smoke regions of a wide range of sizes, colors, textures, and opacity.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 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".