You are on the test server.

Project Details:

Seed composition estimation from in-season standing crop and terminal seed yield using non-invasive imaging

Parent Project: This is the first year of this project.
Checkoff Organization:United Soybean Board
Organization Project Code:2120-152-0201
Project Year:2021
Lead Principal Investigator:Vasit Sagan (St. Louis University)
Co-Principal Investigators:
Felix Fritschi (University of Missouri)

Contributing Organizations

Funding Institutions

Information and Results

Click a section heading to display its contents.

Project Summary

Project Objectives

Project Deliverables

Progress of Work

Updated February 27, 2021:
Given the importance of seed composition for the soybean value chain, we proposed to estimate it based on sensor data and imagery collected in the field (ground- and UAV-based approaches), and post-harvest in the lab. First, we will develop novel machine learning and data fusion algorithms to predict phenotypic traits and estimate seed quality parameters based on in-season, UAV-based sensing modalities and data fusion algorithms. Then, we will establish approaches to estimate seed quality parameters for which accurate non-destructive techniques remain elusive. Finally, we will test the approaches’ robustness for identifying genetic markers for breeding programs. Non-destructive in-season and post-harvest assessment of seed composition will be of great value to stem the decline in seed quality, ensuring U.S. soybeans’ value and competitiveness.

Final Project Results

Benefit to Soybean Farmers

Performance Metrics

Project Years