MDCFT-HSM: Modality-Shift-Aware RGB-T Small-Object Detection for Low-Altitude UAV Remote Sensing
Tianchen Long, Yanwen Wang, Hezhuo Yuan, Qian Zhang, Hongqing Ma, Rijin Zhou, Zhen Wang, Feng Wang · 2026
Read MDCFT-HSM: Modality-Shift-Aware RGB-T Small-Object Detection for Low-Altitude UAV Remote Sensing by Tianchen Long, Yanwen Wang, Hezhuo Yuan, Qian Zhang, Hongqing Ma, Rijin Zhou, Zhen Wang, Feng Wang (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/pmc-13517559.txt.
Description
Tianchen Long and seven co-authors improve drone spotting of small objects in paired color and heat images, in this 2026 paper. Low-flying drones use both normal cameras and thermal sensors, but the two streams are unevenly reliable, and tiny targets fade as features pass through a network. Their method picks a lead stream and guards it.
The design lets the weaker stream in only through small, zero-started additions at four stages, so bad data cannot swamp the lead. Extra blocks sharpen edges and textures, rebuild detail across scales, and reweight features before detection. On two drone benchmarks it scored well, at 49.53 and 83.60 on a standard accuracy scale, while staying small and fast at sixty frames per second. Stress tests showed it tolerates a missing color stream but not severe heat-image damage.
Catalog record
| CREATOR | Tianchen Long, Yanwen Wang, Hezhuo Yuan, Qian Zhang, Hongqing Ma, Rijin Zhou, Zhen Wang, Feng Wang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |