.png){kind=icon}
.png){kind=icon}
Return
Physics-constrained training-free network for low-sampling correlation imaging
.png){kind=icon}
PRE
AI
DOI：10.1016/j.optlastec.2026.114965.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Physics-constrained training-free reconstruction for correlated (ghost) imaging. • The differentiable forward model enables closed-loop, measurement-consistent optimization. • Directly maps low-dimensional bucket signals to high-resolution images without pre-training. • Robust at ultra-low sampling and noise; validated through simulations and optical experiments. • Adapts to new measurement patterns and imaging conditions without retraining.
Keywords:
correlated imaging
physics-constrained
training-free
low-sampling
reconstruction
Journal
O
IF:
0
Papers:
880
Citations:
0