Table 1 Segmentation quality

From: Analyzing microstructure relationships in porous copper using a multi-method machine learning-based approach

Segmentation method

Jaccard Index

Precision

Recall

Accuracy

Pore

Copper

Pore

Copper

Pore

Copper

CTA

0.69

0.81

0.98

0.81

0.70

0.99

0.87

Hybrid

0.77

0.82

0.83

0.93

0.92

0.88

0.89

U-Net CTA

0.59

0.77

0.98

0.77

0.60

0.99

0.83

U-Net Hybrid

0.86

0.89

0.89

0.96

0.96

0.92

0.94

  1. We compare several metrics, e.g., Jaccard Index, precision, recall, and accuracy to validate the segmentation results for the pore and copper. For the segmentation, we use the CTA, the hybrid algorithm as well as the U-Net predictions trained by the CTA and the hybrid algorithm.