High throughput phenotyping in soybean breeding using RGB image vegetation indices based on drone

Area of Science:

• Agricultural Science
• Plant Breeding
• Remote Sensing

Background:

• Optimizing soybean breeding requires efficient phenotyping methods.
• High-throughput phenotyping (HTP) offers a promising avenue for genetic improvement.
• Unmanned aerial vehicles (UAVs) integrated with remote sensing provide novel data acquisition capabilities.

Purpose of the Study:

• To evaluate the effectiveness of HTP using UAV-based RGB imagery for assessing vegetation indices (VIs) in soybean.
• To correlate VIs at different developmental stages with key agronomic performance traits.
• To determine the optimal timing for data collection in HTP for soybean genotype selection.

Main Methods:

• Collected RGB images from UAVs across multiple soybean developmental stages for 60 pure lines.
• Calculated various vegetation indices (VIs) from the imagery.
• Employed linear mixed effects models (REML/BLUP) to analyze genetic variation and predict genotypic values.
• Correlated VIs with agronomic traits such as plant height and maturity.

Main Results:

• Significant genetic differences were observed among soybean genotypes for all evaluated agronomic traits.
• High accuracy and heritability were found for plant height, maturity at R8, and 100-seed weight.
• Vegetation indices showed stronger correlations with plant height and maturity in the later growth stages (105-120 days after sowing).
• Genotype x flight data interactions highlighted the critical importance of data collection timing.

Conclusions:

• HTP using UAV-based VIs is effective for early and late soybean development stages.
• The temporal dimension of HTP is crucial for optimizing soybean genotype selection.
• This methodology refines breeding strategies and agricultural management for soybeans.