Table 3
Training data sets.
| Dataset | Busbar intensity variation (Bx) | Training images | Validation images | Network types | ||
|---|---|---|---|---|---|---|
| S = Simulation data M = Measured MFI data |
S | M | S | M | ||
| BB detection Low variation Simulation data only |
38% | 50 | 0 | 25 | 0 | Instance segmentation/Object detection |
| BB detection Low Variation Simulation + MFI data |
38% | 50 | 7 | 25 | 4 | Instance segmentation/Object detection |
| BB detection High variation Simulation data only |
83% | 50 | 0 | 25 | 0 | Instance segmentation/Object detection |
| BB detection High variation Simulation + MFI data |
83% | 50 | 7 | 25 | 4 | Instance segmentation/Object detection |
| BB defect detection Simulation data only |
38% | 55 | 0 | 27 | 0 | Instance segmentation |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.