A convolutional neural network performs modal decomposition of Hermite-Gaussian beams from one intensity photograph to recover both amplitudes and phases.
In every training epoch, 50000 beam intensity images with a resolution of 128 ×128 are randomly generated to train the CNN
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.optics 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Preliminary study on the modal decomposition of Hermite Gaussian beams via deep learning
A convolutional neural network performs modal decomposition of Hermite-Gaussian beams from one intensity photograph to recover both amplitudes and phases.