Sign in Subscribe
Machine learning

Novel Detection and Classification Method for Shiitake Mushrooms Achieves High Accuracy

Strategic Insights

— 1 min read

Researchers from the Shandong Academy of Agricultural Sciences have developed a novel detection and classification method for shiitake mushrooms using the mamba-YOLO algorithm. This method enables the automatic detection and quality grading of shiitake mushrooms, adhering to the picking standards and grade specifications. Experiments conducted on a self-constructed shiitake mushroom dataset demonstrated the method's high accuracy in precision, recall, and mean average precision.

Key Takeaways:

  • The mamba-YOLO method achieves precision (P), recall (R), and mean average precision (mAP) of 98.89%, 98.79%, and 97.86%, respectively, at an IoU threshold of 50%.
  • The classification accuracies for various categories of shiitake mushrooms range from 96.2% to 98.9%, indicating the method's effectiveness in determining maturity and categorizing mushrooms based on cap texture characteristics.
  • The network detection speed of 8.3 ms is within the acceptable range for real-time applications, and the model's parameters are compact at 6.1 M, facilitating easy deployment and scalability.
  • The lightweight design, precise detection accuracy, and efficient detection speed of mamba-YOLO provide robust technical support for shiitake mushroom harvesting robots.
  • The research was conducted by a team of researchers from the Shandong Academy of Agricultural Sciences, led by Zhen Yang, and including Kangkang Qi, Yangyang Fan, Hualu Song, Zhichao Liang, Shuai Wang, and Fengyun Wang.

Statistics:

  • Precision (P): 98.89%
  • Recall (R): 98.79%
  • Mean Average Precision (mAP): 97.86%
  • Classification accuracy for mushroom stick: 98.1%
  • Classification accuracy for plane-surface immature: 98.3%
  • Classification accuracy for plane-surface mature: 98.2%
  • Classification accuracy for cracked-surface immature: 98.8%
  • Classification accuracy for cracked-surface mature: 98.5%
  • Classification accuracy for deformed mature: 96.2%
  • Classification accuracy for deformed immature: 96.9%
  • Network detection speed: 8.3 ms
  • Model parameter size: 6.1 M

Sources:

  • Detection and classification of Shiitake mushroom fruiting bodies based on Mamba YOLO. Scientific Reports, 2025;15(1):15214.
  • Shandong Academy of Agricultural Sciences
  • Nature Portfolio
  • Nature Publishing Group (www.nature.com/)
  • Scientific Reports (www.nature.com/srep/)

Read more