Location: Agricultural Genetic Resources Preservation Research
Project Number: 3012-21000-017-031-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: Aug 1, 2026
End Date: Jul 31, 2028
Objective:
This award is focused on the development of artificial intelligence (AI)-driven models in seed science to enhance seed quality assessment. Grand Farm will use image analysis to explore machine learning models to predict the viability of seed that have been stored at optimal conditions (-18C) and suboptimal storage conditions (5C).
Approach:
Grand Farm will receive seed from ARS that has been stored for a variety of years at both -18c and 5C. Seed will be planted by Grand Farm and germination will be evaluated in the field, taking weekly drone imagery and traditional crop data, such as growth stage timing, emergence, and stand count. Following the first season, seed image analysis will be conducted, taking images of each side of the seed for each of the accessions evaluated in the field. Grand Farm will perform image analyses and AI model applications to determine if such applications can be used to predict seed health. Once models are developed ARS scientists will test the models with same and different crops, verifying the predictions with viability tests before implementing such model into a routine workflow.