YOLOSeg with applications to wafer die particle defect segmentation
Abstract This study develops the you only look once segmentation (YOLOSeg), an end-to-end instance segmentation model, with applications to segment small particle defects embedded on a wafer die. YOLOSeg uses YOLOv5s as the basis and extends a UNet-like structure to form the segmentation head. YOLOS...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-86323-1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594700288131072 |
|---|---|
| author | Yen-Ting Li Yu-Cheng Chan Chen-Che Huang Yu-Chang Hsu Ssu-Han Chen |
| author_facet | Yen-Ting Li Yu-Cheng Chan Chen-Che Huang Yu-Chang Hsu Ssu-Han Chen |
| author_sort | Yen-Ting Li |
| collection | DOAJ |
| description | Abstract This study develops the you only look once segmentation (YOLOSeg), an end-to-end instance segmentation model, with applications to segment small particle defects embedded on a wafer die. YOLOSeg uses YOLOv5s as the basis and extends a UNet-like structure to form the segmentation head. YOLOSeg can predict not only bounding boxes of particle defects but also the corresponding bounding polygons. Furthermore, YOLOSeg also attempts to obtain a set of better weights by combining with several training tricks such as freezing layers, switching mask loss, using auto-anchor and introducing denoising diffusion probabilistic models (DDPM) image augmentation. The experiment results on the testing image set show that YOLOSeg’s average precision (AP) and intersection over union (IoU) are as high as 0.821 and 0.732 respectively. Even when the sizes of particle defects are extremely small, the performance of YOLOSeg is far superior to current instance segmentation models such as mask R-CNN, YOLACT, YUSEG, and Ultralytics’s YOLOv5s-segmentation. Additionally, preparing the training image set for YOLOSeg is time-saving because it needs neither to collect a large number of defective samples, nor to annotate pseudo defects, nor to design hand-craft features. |
| format | Article |
| id | doaj-art-803aa049400e49ea86e81ece9c795799 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-803aa049400e49ea86e81ece9c7957992025-01-19T12:24:02ZengNature PortfolioScientific Reports2045-23222025-01-0115111410.1038/s41598-025-86323-1YOLOSeg with applications to wafer die particle defect segmentationYen-Ting Li0Yu-Cheng Chan1Chen-Che Huang2Yu-Chang Hsu3Ssu-Han Chen4Circle AI IncorporationCenter for Artificial Intelligence & Data Science, Ming Chi University of TechnologyCathay Financial Holdings Company LimitedDepartment of Industrial Engineering and Management, Ming Chi University of TechnologyDepartment of Industrial Engineering and Management, Ming Chi University of TechnologyAbstract This study develops the you only look once segmentation (YOLOSeg), an end-to-end instance segmentation model, with applications to segment small particle defects embedded on a wafer die. YOLOSeg uses YOLOv5s as the basis and extends a UNet-like structure to form the segmentation head. YOLOSeg can predict not only bounding boxes of particle defects but also the corresponding bounding polygons. Furthermore, YOLOSeg also attempts to obtain a set of better weights by combining with several training tricks such as freezing layers, switching mask loss, using auto-anchor and introducing denoising diffusion probabilistic models (DDPM) image augmentation. The experiment results on the testing image set show that YOLOSeg’s average precision (AP) and intersection over union (IoU) are as high as 0.821 and 0.732 respectively. Even when the sizes of particle defects are extremely small, the performance of YOLOSeg is far superior to current instance segmentation models such as mask R-CNN, YOLACT, YUSEG, and Ultralytics’s YOLOv5s-segmentation. Additionally, preparing the training image set for YOLOSeg is time-saving because it needs neither to collect a large number of defective samples, nor to annotate pseudo defects, nor to design hand-craft features.https://doi.org/10.1038/s41598-025-86323-1Auto-annotationDefect segmentationWafer dieYou only look once (YOLO)Denoising diffusion probabilistic models (DDPM) |
| spellingShingle | Yen-Ting Li Yu-Cheng Chan Chen-Che Huang Yu-Chang Hsu Ssu-Han Chen YOLOSeg with applications to wafer die particle defect segmentation Scientific Reports Auto-annotation Defect segmentation Wafer die You only look once (YOLO) Denoising diffusion probabilistic models (DDPM) |
| title | YOLOSeg with applications to wafer die particle defect segmentation |
| title_full | YOLOSeg with applications to wafer die particle defect segmentation |
| title_fullStr | YOLOSeg with applications to wafer die particle defect segmentation |
| title_full_unstemmed | YOLOSeg with applications to wafer die particle defect segmentation |
| title_short | YOLOSeg with applications to wafer die particle defect segmentation |
| title_sort | yoloseg with applications to wafer die particle defect segmentation |
| topic | Auto-annotation Defect segmentation Wafer die You only look once (YOLO) Denoising diffusion probabilistic models (DDPM) |
| url | https://doi.org/10.1038/s41598-025-86323-1 |
| work_keys_str_mv | AT yentingli yolosegwithapplicationstowaferdieparticledefectsegmentation AT yuchengchan yolosegwithapplicationstowaferdieparticledefectsegmentation AT chenchehuang yolosegwithapplicationstowaferdieparticledefectsegmentation AT yuchanghsu yolosegwithapplicationstowaferdieparticledefectsegmentation AT ssuhanchen yolosegwithapplicationstowaferdieparticledefectsegmentation |