Search alternatives:
ap optimization » ai optimization (Expand Search), art optimization (Expand Search), gpu optimization (Expand Search)
binary basic » binary mask (Expand Search)
basic global » based global (Expand Search), waqi global (Expand Search)
less based » lens based (Expand Search), lemos based (Expand Search), degs based (Expand Search)
based ap » based _ (Expand Search), based 3d (Expand Search), based co (Expand Search)
ap optimization » ai optimization (Expand Search), art optimization (Expand Search), gpu optimization (Expand Search)
binary basic » binary mask (Expand Search)
basic global » based global (Expand Search), waqi global (Expand Search)
less based » lens based (Expand Search), lemos based (Expand Search), degs based (Expand Search)
based ap » based _ (Expand Search), based 3d (Expand Search), based co (Expand Search)
-
1
Schematic diagram of weld surface defects.
Published 2024“…The experimental results show that the defect detection <a href="mailto:mAP@0.5" target="_blank">mAP@0.5</a> based on the improved YOLOv7 algorithm can reach 72.2%, which is 11% higher than that of YOLOv7, and the model calculation amount and parameter amount are reduced by 75.6% and 60.3%, respectively. …”
-
2
Improved YOLOv7 network structure.
Published 2024“…The experimental results show that the defect detection <a href="mailto:mAP@0.5" target="_blank">mAP@0.5</a> based on the improved YOLOv7 algorithm can reach 72.2%, which is 11% higher than that of YOLOv7, and the model calculation amount and parameter amount are reduced by 75.6% and 60.3%, respectively. …”
-
3
Renderings of data enhancements.
Published 2024“…The experimental results show that the defect detection <a href="mailto:mAP@0.5" target="_blank">mAP@0.5</a> based on the improved YOLOv7 algorithm can reach 72.2%, which is 11% higher than that of YOLOv7, and the model calculation amount and parameter amount are reduced by 75.6% and 60.3%, respectively. …”
-
4
Number and size of marked defects.
Published 2024“…The experimental results show that the defect detection <a href="mailto:mAP@0.5" target="_blank">mAP@0.5</a> based on the improved YOLOv7 algorithm can reach 72.2%, which is 11% higher than that of YOLOv7, and the model calculation amount and parameter amount are reduced by 75.6% and 60.3%, respectively. …”
-
5
Loss function curve.
Published 2024“…The experimental results show that the defect detection <a href="mailto:mAP@0.5" target="_blank">mAP@0.5</a> based on the improved YOLOv7 algorithm can reach 72.2%, which is 11% higher than that of YOLOv7, and the model calculation amount and parameter amount are reduced by 75.6% and 60.3%, respectively. …”
-
6
Precision-Recall curve.
Published 2024“…The experimental results show that the defect detection <a href="mailto:mAP@0.5" target="_blank">mAP@0.5</a> based on the improved YOLOv7 algorithm can reach 72.2%, which is 11% higher than that of YOLOv7, and the model calculation amount and parameter amount are reduced by 75.6% and 60.3%, respectively. …”
-
7
Comparison experiment results.
Published 2024“…The experimental results show that the defect detection <a href="mailto:mAP@0.5" target="_blank">mAP@0.5</a> based on the improved YOLOv7 algorithm can reach 72.2%, which is 11% higher than that of YOLOv7, and the model calculation amount and parameter amount are reduced by 75.6% and 60.3%, respectively. …”
-
8
Ablation experiment results.
Published 2024“…The experimental results show that the defect detection <a href="mailto:mAP@0.5" target="_blank">mAP@0.5</a> based on the improved YOLOv7 algorithm can reach 72.2%, which is 11% higher than that of YOLOv7, and the model calculation amount and parameter amount are reduced by 75.6% and 60.3%, respectively. …”
-
9
Deepwise separable convolution structure diagram.
Published 2024“…The experimental results show that the defect detection <a href="mailto:mAP@0.5" target="_blank">mAP@0.5</a> based on the improved YOLOv7 algorithm can reach 72.2%, which is 11% higher than that of YOLOv7, and the model calculation amount and parameter amount are reduced by 75.6% and 60.3%, respectively. …”
-
10
Data Sheet 1_Leveraging automated time-lapse microscopy coupled with deep learning to automate colony forming assay.docx
Published 2025“…Brightfield images were used to train a YOLOv8 object detection network, achieving a mAP50 score of 86% for identifying single cells, clusters, and colonies, and 97% accuracy for Z-stack colony identification with a multi-object tracking algorithm. …”