Table 4 Accuracy of the baseline model’s object detection and image segmentation tasks with DynamicDA
From: Pest detection in dynamic environments: an adaptive continual test-time domain adaptation strategy
DynamicDA in original environment | Precision\(\uparrow\) | Recall\(\uparrow\) | mAP50\(\uparrow\) | mAP50-95\(\uparrow\) |
|---|---|---|---|---|
YOLOv9c | 0.714 | 0.753 | 0.768 | 0.511 |
YOLOv9e | 0.7 | 0.73 | 0.764 | 0.52 |
YOLOv9c-seg | 0.658 | 0.683 | 0.651 | 0.527 |
YOLOv9e-seg | 0.668 | 0.635 | 0.644 | 0.533 |
DynamicDA in target environment | Precision\(\uparrow\) | Recall\(\uparrow\) | mAP50\(\uparrow\) | mAP50-95\(\uparrow\) |
|---|---|---|---|---|
YOLOv9c | 0.502 | 0.478 | 0.494 | 0.289 |
YOLOv9e | 0.529 | 0.562 | 0.541 | 0.334 |
YOLOv9c-seg | 0.528 | 0.421 | 0.454 | 0.328 |
YOLOv9e-seg | 0.533 | 0.438 | 0.459 | 0.344 |