Breakthrough in Corn Seed Detection: A Lightweight Model for Accurate Phenotyping

Strategic Insights -

Researchers at Hainan University, Haikou, People's Republic of China, have developed a novel lightweight corn seed kernel rapid detection model based on YOLOv11n (LWCD-YOLO). This innovative approach is poised to revolutionize the field of agriculture, enabling precise and rapid detection of corn seeds. Funded by the National Key R&D Program of China and the National Talent Foundation Project of China, the study proposes a new module design and optimization techniques to achieve high detection performance while minimizing model complexity.

Key Takeaways:

  • The LWCD-YOLO model reduces model complexity by 50% and parameter count by 51% compared to the original YOLOv11n.
  • The model achieves a precision of 99.978%, mean Average Precision at 0.50 of 99.491%, and mean Average Precision at 0.50:0.95 of 99.262%.
  • The model requires 1.27 million parameters and 3.5 G of FLOPs, resulting in a 94% improvement in FPS.
  • The model's performance is superior to current mainstream object detection models, making it suitable for fast and precise detection of corn seeds.
  • The research team, led by Wenbin Sun, includes Kang Xu, Dongquan Chen, Danyang Lv, Ranbing Yang, Songmei Yang, Rong Wang, Ling Wang, and Lu Chen.

Statistics:

  • The model's precision is 99.978%.
  • The mean Average Precision at 0.50 is 99.491%.
  • The mean Average Precision at 0.50:0.95 is 99.262%.
  • The model size is reduced by 50%.
  • The parameter count is reduced by 51%.
  • The computational complexity is reduced by 44%.
  • FPS is improved by 94%.

Sources:

  • Agriculture, Volume 15, Issue 18, 2025, Article No. 1968, "LWCD-YOLO: A Lightweight Corn Seed Kernel Fast Detection Algorithm Based on YOLOv11n", available at http://www.mdpi.com/journal/agriculture.
  • DOI: 10.3390/agriculture15181968.
  • The publisher for Agriculture is MDPI AG.