Claim · model-written summary
TIoU-YOLOv8n-Pruned reached 94.1% accuracy and 57.2% mAP@0.5:0.95, outperforming YOLOv5n, YOLOv8n, YOLOv11n and other YOLO variants.
- Taxon
- Walnutt0
- Outcome
- accuracy and mAP@0.5:0.95e0
- Direction
- outperforminge0
- Value
- 94.1% and 57.2%e0
- Comparator
- YOLOv5n, YOLOv8n, YOLOv11ne0
Not reportedMaterialGeneCompoundOrgan / stageInteractorInteractor roleSecond interactorSecond interactor roleTreatmentSettingQualifier
Verbatim
t0“Walnut Trunks”
e0“while its accuracy and mAP@0.5:0.95 reach 94.1% and 57.2%, outperforming YOLOv5n, YOLOv8n, YOLOv11n”
Title
Rapid Identification and Accurate Localization of Walnut Trunks Based on TIoU-YOLOv8n-Pruned
Chenchen Ye; Yan Xu; Jianping Zhou; et al. Agriculture 2025. doi:10.3390/agriculture15232405
Other claims from this paper (5)
- Method / resource TIoU-YOLOv8n-Pruned has 950,000 parameters.
- Method / resource TIoU-YOLOv8n-Pruned requires 3.6 GFLOPs.
- Method / resource Channel pruning removed 55% of the model non-essential channels.
- Intervention effect Trajectory fitting and coordinate correction significantly reduced vibration-induced errors in walnut trunk localization.
- Intervention effect Trajectory fitting and coordinate correction enhanced localization accuracy of walnut trunks.