MSA-DET: A Multi-Scale Attention Network with Adaptive Feature Fusion for SAR Ship Detection
Sai Wan, Zhiyong Tao, Lu Chen · 2026
Read MSA-DET: A Multi-Scale Attention Network with Adaptive Feature Fusion for SAR Ship Detection by Sai Wan, Zhiyong Tao, Lu Chen (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/PMC13364323.txt.
Description
Sai Wan and co-workers present MSA-DET in 2026, a net for finding ships in radar images from space. Radar views carry speckle noise, messy coasts, and ships of vastly unlike size. Their design adds focus blocks that scan in two ways at once and a head that blends scale views by learned weights.
On two ship data sets the net beat its base model by clear gaps on both strict and loose scores. The cost was a larger model and more compute. The trade may suit coast watch work, where missed small ships carry real risk.
Catalog record
| CREATOR | Sai Wan, Zhiyong Tao, Lu Chen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43148 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |