Estimation of a New Canopy Structure Parameter for Rice Using Smartphone Photography

Area of Science:

• Agricultural Science
• Remote Sensing
• Plant Physiology

Background:

• Accurate rice yield estimation is crucial for food security.
• Traditional methods for monitoring crop growth are often labor-intensive and time-consuming.
• Digital imaging offers a potential low-cost, rapid alternative for agricultural monitoring.

Purpose of the Study:

• To develop a low-cost method for obtaining rice growth information using smartphone digital images.
• To introduce and validate a new canopy volume parameter (CVP) for characterizing rice yield.
• To establish a predictive model for rice yield using image-derived parameters.

Main Methods:

• Acquisition of rice canopy images using a smartphone.
• Extraction of image feature parameters: canopy cover (CC) and vegetation indices (VIs).
• Development of a random forest (RF) regression model to predict CVP using CC and VIs.

Main Results:

• The proposed canopy volume parameter (CVP) demonstrated superior performance over leaf area index (LAI) and plant height (PH) in predicting final rice yield.
• A local modeling approach for CVP prediction, distinguishing rice varieties, achieved the highest accuracy (R² = 0.92, RMSE = 0.44).
• The developed random forest model effectively predicted rice yield using smartphone-derived image features.

Conclusions:

• Smartphone-based digital imaging provides a viable and efficient tool for tracking crop growth.
• The canopy volume parameter (CVP) is a robust indicator for estimating rice yields.
• This approach offers technical support for rapid crop monitoring and yield prediction in agriculture.