Table 8 Metrics for each pathology type (YOLOv9e-seg)
From: Deep learning-driven pathology detection and analysis in historic masonry buildings of Suzhou
Class | Images | Instances | Box | Mask | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
Precision | Recall | mAP50 | mAP50-95 | Precision | Recall | mAP50 | mAP50-95 | |||
all | 174 | 1963 | 0.519 | 0.351 | 0.407 | 0.238 | 0.382 | 0.423 | 0.383 | 0.217 |
ML | 3 | 23 | 0.794 | 0.217 | 0.563 | 0.294 | 0.819 | 0.394 | 0.599 | 0.34 |
DD | 148 | 1001 | 0.461 | 0.46 | 0.433 | 0.304 | 0.316 | 0.59 | 0.415 | 0.27 |
CR | 59 | 144 | 0.528 | 0.451 | 0.44 | 0.239 | 0.31 | 0.484 | 0.419 | 0.183 |
SS | 137 | 746 | 0.579 | 0.383 | 0.454 | 0.306 | 0.415 | 0.504 | 0.444 | 0.282 |
BI | 20 | 49 | 0.231 | 0.245 | 0.146 | 0.0452 | 0.0514 | 0.143 | 0.0403 | 0.0122 |