Early diagnosis of the physiological quality of soybean seeds: a new frontier in the production system in times of uncertainty climate

• Grant number: 24/06710-4
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: December 01, 2024
• End date: November 30, 2026
• Field of knowledge: Agronomical Sciences - Agronomy - Crop Science
• Principal Investigator: Clíssia Barboza Mastrangelo
• Grantee: Thiago Barbosa Batista
• Host Institution: Centro de Energia Nuclear na Agricultura (CENA). Universidade de São Paulo (USP). Piracicaba , SP, Brazil

Abstract

Climate uncertainties place production fields in vulnerable conditions to obtain seeds with high physiological potential since stresses during maturation can reduce the physiological quality of seeds. While the problem occurs, in theory, it would be possible diagnose it, and thus, predict that a field will not produce seeds in the standards required for commercial class. This "early diagnosis" has high potential in the seed industry, as it would provide producers with logistical and strategic information. In this project, the proposal is to develop a methodology for early diagnosis of quality physiological characteristics of soybean seeds during maturation produced under stressful conditions. A methodology will use data from multispectral images of seeds, particularly parameters of color, shape, texture, reflectance and autofluorescence, and validation will be done using artificial intelligence algorithms, with autonomous prediction of loss of physiological quality of the seeds. In this sense it will expect to validate a robust methodology using multispectral images integrated with machine learning for early diagnosis of the physiological quality of seeds of soybeans, which will contribute to the implementation of innovative approaches in the system of seed production.