Research on Wheat Spike Detection and Counting Methods Based on Lightweight Improved Network
This article presents an innovative study on the use of lightweight methods to improve the YOLOv5s network. It introduces two model lightweight improvement methods: inverted residual structure and Ghost Bottleneck module. Based on the improved models, lightweight improvement comparison experiments and model comparison experiments are designed, and the experimental results are discussed and analyzed.
Authors
- Miao Yu (ORCID: https://orcid.org/0000-0003-3482-3489)
- Yu Wang (ORCID: https://orcid.org/0009-0005-0661-311X)
Institutions
- Liaoning Academy of Agricultural Sciences (CN)
Publication Details
- Published
- 2024-08-29
- DOI
- https://doi.org/10.1109/icsece61636.2024.10729443
- Primary Topic
- Smart Agriculture and AI
- Type
- article
- Field-Weighted Citation Impact
- 0.00