ASCL.net

Astrophysics Source Code Library

Making codes discoverable since 1999

ASCL Code Record

[ascl:2503.039] R2D2: Residual-to-Residual DNN series for high-Dynamic range imaging

R2D2 (Residual-to-Residual DNN series for high-Dynamic range imaging) performs synthesis imaging for radio interferometry. The R2D2 algorithm takes a hybrid structure between a Plug-and-Play (PnP) algorithm and a learned version of the well-known Matching Pursuit algorithm. Its reconstruction is formed as a series of residual images, iteratively estimated as outputs of iteration-specific Deep Neural Networks (DNNs), each taking the previous iteration’s image estimate and associated back-projected data residual as inputs. The primary application of the R2D2 algorithm is to solve large-scale high-resolution high-dynamic range inverse problems in radio astronomy, more specifically 2D planar monochromatic intensity imaging.

Code site:
https://github.com/basp-group/R2D2-SII
Described in:
https://ui.adsabs.harvard.edu/abs/2024ApJS..273....3A https://ui.adsabs.harvard.edu/abs/2024ApJ...966L..34D
Bibcode:
2025ascl.soft03039A

Views: 51

ascl:2503.039
Add this shield to your page
Copy the above HTML to add this shield to your code's website.