Smart Buildings: Water Leakage Detection Using TinyML
Othmane Atanane, Asmaa Mourhir, Nabil Benamar, Marco Zennaro, Luis Hernández-Callejo · 2023
Read Smart Buildings: Water Leakage Detection Using TinyML by Othmane Atanane, Asmaa Mourhir, Nabil Benamar, Marco Zennaro, Luis Hernández-Callejo (2023) 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-10675406.txt.
Description
Othmane Atanane and colleagues built a tiny learning system in 2023 to catch water leaks. Cities face rising water stress and shortages. Leaks in building pipes waste large amounts silently. Sensors plus on-device learning can spot leaks locally. Local decisions cut delay and central system costs. The team used an existing sound dataset for plastic pipes. Sound clips became image-like patterns for analysis. Five image network designs were adapted through transfer learning. The best model was then shrunk for a small board.
All networks learned to detect leak sounds well. One efficient design scored highest on every measure. Its test accuracy came close to ninety-eight percent. After compression, it ran on an Arduino-class edge device. Inference took about two seconds per check. Memory and storage needs stayed very small. The authors present a near hands-off leak watcher. Real buildings could host such devices widely, they argue.
Catalog record
| CREATOR | Othmane Atanane, Asmaa Mourhir, Nabil Benamar, Marco Zennaro, Luis Hernández-Callejo |
|---|---|
| YEAR | 2023 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |