A hybrid multi-scale indirect vision detector algorithm for heritage building fire detection
Jian Liang, Junsheng Cheng · 2025
Read A hybrid multi-scale indirect vision detector algorithm for heritage building fire detection by Jian Liang, Junsheng Cheng (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12800248.txt.
Description
Jian Liang and Junsheng Cheng present a two thousand twenty-five study of a vision algorithm for spotting fires in heritage buildings. Mirrors give indirect views around fragile interiors, but flames appear tiny and views uneven. Standard detectors miss such small targets.
Their hybrid algorithm rebalances brightness across mirror views first. Parallel feature extraction at several scales catches flames of different sizes. A fusion network then mixes detail across levels. On a purpose-built heritage fire dataset, accuracy beat a leading detector baseline on both measures used. All gains came while staying lightweight enough for practical deployment.
Catalog record
| CREATOR | Jian Liang, Junsheng Cheng |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 74 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |