Fig. 5: Recovered physical parameters.
From: Inverse imaging with elastic waves driven by unsupervised machine learning
a–f corresponds to the recovered Young’s modulus \({E}_{0}\) and heights of three blocks \({h}_{1}\), \({h}_{2}\), and \({h}_{3}\) of the six 3D-printed samples I, II–VI from experimental data by the ML model. For each panel, the first row of the bar plot shows the predictions from experimental data, the second row is from the simulation data, and the third row is the ground truth.
