Schematic diagram of weld surface defects. by Xiangqian Xu (17310895)

“…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. …”

Improved YOLOv7 network structure. by Xiangqian Xu (17310895)

“…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. …”

Renderings of data enhancements. by Xiangqian Xu (17310895)

“…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. …”

Number and size of marked defects. by Xiangqian Xu (17310895)

“…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. …”

Loss function curve. by Xiangqian Xu (17310895)

“…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. …”

Precision-Recall curve. by Xiangqian Xu (17310895)

“…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. …”

Comparison experiment results. by Xiangqian Xu (17310895)

“…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. …”

Ablation experiment results. by Xiangqian Xu (17310895)

“…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. …”

Deepwise separable convolution structure diagram. by Xiangqian Xu (17310895)

“…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. …”

Data Sheet 1_Leveraging automated time-lapse microscopy coupled with deep learning to automate colony forming assay.docx by Anusha Klett (20748443)

“…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. …”