Fig. 1From: Machine learning as a tool for classifying electron tomographic reconstructionsReconstruction artefacts raw (a) and WBP-reconstructed versions of a modified Shepp-Logan phantom (b) with their histograms shown in (c) and (d), respectively. Inset regions in (b) are ×2 magnified portions of the lower central part—left is from the source phantom, right is from the reconstruction. Absolute intensity values should not be directly compared, as the reconstruction algorithm automatically scales the output data to saturate a 0—255 range. The contrast in the right inset of (b) has been enhanced for the purpose of improved visibilityBack to article page