Fig. 7
From: Enhancing image resolution of confocal fluorescence microscopy with deep learning
Super-resolution imaging of confocal images captured with the Nikon A1R MP+ microscope using TCAN. (a, d, g, j) Diffraction-limited confocal microscopy images of the microtubules are used as input to the network. (b, e) The Super-resolved network inference images before downsampling the input images to match the pixel size of the training data. (c, f) The super-resolved network inference images after downsampling the input images to match the pixel sizes of the training data. (h, k) The Super-resolved network inference images without the need for downsampling the input images to match the pixel size of the training data. (i, l) Intensity profiles along the white dashed lines and blue solid lines in different images. Scale bar in (c), (f), (h) and (k) are 6 μm, 3 μm, 6 μm and 10 μm, respectively