Publication: Detection of solid wood defect using an improved yolov7 algorithm
| datacite.subject.fos | oecd::Engineering and technology::Electrical engineering, Electronic engineering, Information engineering::Electrical and electronic engineering | |
| dc.contributor.author | Sui, Yun Hen | |
| dc.date.accessioned | 2025-03-27T08:22:34Z | |
| dc.date.available | 2025-03-27T08:22:34Z | |
| dc.date.issued | 2023-08 | |
| dc.description.abstract | Conventional way of wood defect detection methods has a lot of drawbacks such as low accuracy, human dependent, and low speed. Deep learning defect detection for solid wood panel is therefore implemented to improve the accuracy of defect detection and reduce human dependencies. YOLOv7 algorithm is chosen to be the algorithm to be enhanced. In this paper, YOLOv7 is improved by implementing a Bi-directional Feature Pyramid Network (BiFPN) to increase the accuracy. Four types of defects are evaluated in this paper, namely crack, pinhole, dead knot and undersized. The obtained result is compared with the results from the original YOLOv7 and other algorithms to ensure that the objectives are achieved to a certain extend. The result proved that the YOLOv7-BiFPN can identify wood defects with higher accuracy. | |
| dc.identifier.uri | https://erepo.usm.my/handle/123456789/21387 | |
| dc.language.iso | en | |
| dc.title | Detection of solid wood defect using an improved yolov7 algorithm | |
| dc.type | Resource Types::text::report::technical report | |
| dspace.entity.type | Publication | |
| oairecerif.author.affiliation | Universiti Sains Malaysia |