The 3D dynamic visualization simulation of rice plant based on morphological structure model and the application in...

Area of Science:

• Agricultural Science
• Computational Biology
• Plant Science

Background:

• Virtual crops are essential for exploring crop phenotypes in crop modeling.
• Accurate visualization and prediction of plant morphology are crucial for agricultural research.

Purpose of the Study:

• To achieve 3D dynamic visualization of individual rice plants and populations.
• To predict rice phenotype using a virtual rice model.
• To integrate above-ground and root system synchronization for realistic simulations.

Main Methods:

• Developed a 3D dynamic visualization system for rice growth stages.
• Integrated above-ground and root system synchronization in the virtual rice model.
• Employed leaf segmentation within the virtual rice model to predict leaf area index (LAI).

Main Results:

• Achieved realistic 3D dynamic visualizations of rice individuals and populations.
• Predicted LAI with relative errors between 7.58% and 12.69%.
• Evaluation metrics (RMSE=0.56, MAE=0.55, R²=0.86) demonstrated high accuracy.

Conclusions:

• The virtual rice model offers realistic visualization and accurate phenotype prediction capabilities.
• This research provides technical and modeling support for advancing virtual crop development.
• The findings enhance the predictive power for rice morphological changes and LAI.