Answers are grounded in this paper and cite their sources.
Assessing plant nutritional status through phenotypic traits is essential for optimizing fertilization in cut chrysanthemum ( Chrysanthemum morifolium Ramat.) production; however, practical and rapid methods for nitrogen level identification remain limited. In this study, cultivation experiments were conducted on the major cultivar ‘Jingcheng’ under three nitrogen supply levels within typical production ranges. A PSPNet (Pyramid Scene Parsing Network) model was used to segment entire leaves and top lobes, and a YOLO v11n-pose model was employed to detect ten key points per leaf, enabling extraction of ten measured and nine calculated shape parameters. The PSPNet achieved a mean accuracy of 0.9926 with 59.21 GFLOPs, while YOLO v11n-pose obtained a mAP0.5 of 0.982 with only 2.73 M parameters, ensuring efficient and accurate feature extraction. The extracted leaf shape parameters were measured with mean relative errors (MRE) below 2.2%, confirming the accuracy of the proposed morphological parameter extraction framework. Calculated parameters showed higher stability (CV ≈ 10%) than directly measured ones (CV > 20%), particularly in mid- to lower-position leaves. One-way ANOVA indicated that eight calculated parameters of lower and middle leaves differed significantly among nitrogen treatments (P < 0.05). Among tested classifiers, the XGBoost model performed best, with an accuracy of 0.8617 and a Kappa coefficient of 0.7926. Feature importance analysis identified the Ratio of terminal lobe length to leaf length and Top lobe aspect ratio as the key discriminative indicators. The results establish an efficient and interpretable morphological framework for rapid identification of nitrogen supply levels in cut chrysanthemum, providing support for further precision nutrient management and intelligent cultivation.
No takes yet. Share an insight, caveat, or question.
Xi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: