Fig. 6: Artificial neural network for defect classification. | Nature Communications

Fig. 6: Artificial neural network for defect classification.

From: Feature-based volumetric defect classification in metal additive manufacturing

Fig. 6

a Rankings of the morphological parameters in terms of their importance obtained via PFI analysis. b Architecture of the ANN model: only connections with relatively large weights (i.e., large absolute values) in the ANN are shown. I, L1, L2, and O represent input layer, 1st hidden layer, 2nd hidden layer, and output layer of ANN, respectively.

Back to article page