A Canopy Information Measurement Method for Modern Standardized Apple Orchards Based on UAV Multimodal Information

Area of Science:

• Precision Agriculture
• Remote Sensing
• Agricultural Engineering

Background:

• Accurate canopy information is crucial for optimizing management in modern standardized apple orchards.
• Traditional methods for canopy measurement are often labor-intensive and lack efficiency.
• Integrating Unmanned Aerial Vehicle (UAV) technology offers a promising solution for large-scale, detailed orchard monitoring.

Purpose of the Study:

• To propose and validate a novel method for apple orchard canopy information measurement using UAV multimodal data.
• To develop accurate detection techniques for orchard rows and columns based on UAV-derived imagery.
• To establish a yield prediction model leveraging UAV-based canopy morphological and spectral data.

Main Methods:

• Utilized a quadrotor UAV equipped with a visual imaging system to capture orchard canopy data.
• Generated 3D point-cloud models and vegetation index images using Pix4Dmapper software.
• Developed a row and column detection method based on grayscale projection (RCGP) and analyzed canopy morphology (H, SXOY, V) and yield prediction using artificial neural networks.

Main Results:

• The RCGP method achieved 100.00% accuracy for row detection and 98.71-100.00% for column detection.
• UAV-measured canopy morphological parameters (H, SXOY, V) showed high correlation with hand-measured values (R² = 0.91-0.94) with minimal deviation (RADavg = 1.72-7.90%).
• The yield prediction model achieved an R² of 0.83-0.88 and a relative average deviation (RAD) of 8.05-9.76%.

Conclusions:

• The proposed UAV-based method enables accurate remote evaluation of 3D canopy morphology in apple orchards.
• This approach significantly enhances yield prediction capabilities for modern standardized orchards.
• The method contributes to improving orchard informatization and production management efficiency.