EHCTNet: Enhanced Hybrid of CNN and Transformer Network for Remote Sensing Image Change Detection
Junjie Yang, Haibo Wan, Zhihai Shang · 2025
Read EHCTNet: Enhanced Hybrid of CNN and Transformer Network for Remote Sensing Image Change Detection by Junjie Yang, Haibo Wan, Zhihai Shang (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/arxiv-2501-01238.txt.
Description
Yang, Wan, and Shang build EHCTNet in twenty twenty-five, a hybrid network that spots change in satellite image pairs. Missed changes cost more than false alarms in this field. Past tools chased precision and left broken, partial outlines. Their design aims squarely at recall of the changes that matter.
Two branches pull multi scale features from the image pair. Frequency parts of those features get refined in two stages. A Kolmogorov Arnold based module mines the meaning level signal. Tests show fuller, cleaner change regions than top rivals. Neighbor borders came out sharper as well.
Catalog record
| CREATOR | Junjie Yang, Haibo Wan, Zhihai Shang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-01238