Figure 3
From: Multi-resolution convolutional neural networks for inverse problems
Amplitudes and phases predicted from defocused images. First row: 51 defocused Hela cell images recorded at focal planes exponentially spaced from −500 μm to 500 μm (only 11 are shown). (a) Phases (top) and amplitudes (bottom) reconstructed by MFTIE from all 51 images32. A range-of-interest of 470 × 520 pixels is selected; (b) MCNN prediction using 2 images taken from −1 μm and 1 μm; (c) MCNN prediction using 4 images taken from −1.30 μm to 1.30 μm; (d) U-Net prediction using 4 images taken from −1.30 μm to 1.30 μm; (e) CDINN prediction using 4 images taken from − 1.30 μm to 1.30 μm.
