Fig. 3
From: Enhancing image resolution of confocal fluorescence microscopy with deep learning
Super-resolution imaging of the nano-beads and nucleus using TCAN. (a) A diffraction-limited confocal microscopy image of fluorescent nano-beads is used as input to the network. (b) The super-resolved output image. (c) The STED image of the same field of view. (d, j) Examples of closely spaced nano-beads or NPCs that cannot be resolved by confocal microscopy. (e, k) The trained network takes (d) and (j) as input and resolves the individual nano-beads or NPCs. (f, l) The STED microscopy image. (g) A diffraction-limited confocal microscopy image of HeLa cell nucleus is used as input to the network. (h) The super-resolved output image. (i) The STED image of the same field of view. (m, n) Intensity profiles along the white dashed lines and green solidlines in different images. Experiments are repeated with 20 images, achieving similar results. Scale bars, 4 μm