TL-DETR: Efficient transmission line defect detection for edge deployment
Yong Zhang, Runming Zhao · 2026
Read TL-DETR: Efficient transmission line defect detection for edge deployment by Yong Zhang, Runming Zhao (2026) 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/PMC13258123.txt.
Description
Yong Zhang and Runming Zhao wrote a 2026 engineering article on TL-DETR, a computer vision model for spotting defects on power transmission lines. Edge deployment means running the model on small devices near the lines, with limited computing power. Defects vary hugely in size and sit against cluttered backgrounds, which makes detection hard. Their design adds multi-scale feature tools and a dynamic sparse attention system to focus on small defects while cutting computation.
On the CableInspect-ADs dataset, TL-DETR reached precision of 91.4 percent and a detection score called mAP50 of 86.0 percent, both clear of the baseline RT-DETR model's marks. The gains over that baseline were 3.2 points on precision and 2.9 points on the detection score. Further tests showed the model also generalized well to insulators, vibration dampers, and bolts. The authors say it balances accuracy with efficiency well enough for real field use.
Catalog record
| CREATOR | Yong Zhang, Runming Zhao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 72810 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |