Development of a hybrid algorithm to predict room fire flashovers based on vision data
Bibliographic record
Abstract
One of the most deadly situations that firefighters could face in firefighting is flashover, which is sudden fire propagation occurring in a room with all the items in the room bursting into the fire. In general, firefighters need years of training to identify and predict the flashover. Although the decades of experimental and numerical fire research shed light on the room fire dynamics, there are still gaps in transferring the fire science to the fire ground where innovative yet simple solutions are needed to overcome the harsh environment. This project is to develop a robust smart firefighting tool that can be easily deployed like cameras to the fire ground and provide effective assistance to firefighters. One key ability of the smart fighting tool would be assisting firefighters in detecting impending deadly flashovers. An explorative study is conducted adopting deep learning methods in the processing of smoke and flame video images. Scientific knowledge of room fires is also coupled to build an algorithm that requires less hardware but produces high accuracy. The hybrid system combining deep learning methods and fire safety knowledge only requires RGB vision data for flashover prediction, which can be acquired by any camera used by firefighters. The system converts the RGB inputs to thermal images and processes the flashover analysis with images classified as smoke and flame. The system was tested with video data obtained from various fire tests, and the performance was evaluated and compared with other existing models. The hybrid algorithm of the flashover prediction system demonstrated promising performance by surpassing other existing methods designed for similar tasks, with high prediction accuracy
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".