Efficient FPGA-accelerated Convolutional Neural Networks for Cloud Detection on CubeSats
Angela Cratere, \IEEEmembership Student Member, IEEE, M. Salim Farissi, Andrea Carbone, Marcello Asciolla, \IEEEmembership Student Member, IEEE, Maria Rizzi, Francesco Dell’Olio, \IEEEmembership Senior Member, IEEE, Augusto Nascetti, \IEEEmembership Member, IEEE and Dario Spiller · 2025
Read Efficient FPGA-accelerated Convolutional Neural Networks for Cloud Detection on CubeSats by Angela Cratere, \IEEEmembership Student Member, IEEE, M. Salim Farissi, Andrea Carbone, Marcello Asciolla, \IEEEmembership Student Member, IEEE, Maria Rizzi, Francesco Dell’Olio, \IEEEmembership Senior Member, IEEE, Augusto Nascetti, \IEEEmembership Member, IEEE and Dario Spiller (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-2504-03891.txt.
Description
Angela Cratere led a team putting neural networks on small satellites in a 2025 paper. Their task is spotting clouds in Earth images aboard CubeSats, where computing power is tight. Four network models were sped up on a programmable chip using a vendor AI framework. Two models judge single pixels or patches, and two judge whole images.
Trimming channels cut model size and math sharply with minimal accuracy loss. Shrinking numbers to eight-bit form kept accuracy nearly intact on the chip. The two whole-image models ran in real time at low power, beating prior onboard cloud detectors. The authors see chip accelerators widening what small satellites can do.
Catalog record
| CREATOR | Angela Cratere, \IEEEmembership Student Member, IEEE, M. Salim Farissi, Andrea Carbone, Marcello Asciolla, \IEEEmembership Student Member, IEEE, Maria Rizzi, Francesco Dell’Olio, \IEEEmembership Senior Member, IEEE, Augusto Nascetti, \IEEEmembership Member, IEEE and Dario Spiller |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-03891