A Reliable Method to Recognize Soybean Seed Maturation Stages Based on Autofluorescence-Spectral Imaging Combined...

Area of Science:

• Agricultural Science
• Biotechnology
• Spectroscopy

Background:

• Seed quality is crucial for crop performance, necessitating advanced diagnostic methods.
• Autofluorescence-spectral imaging offers a novel approach to assess seed physiological status.
• Identifying seed maturation stages is key to selecting high-quality seeds.

Purpose of the Study:

• To investigate the potential of autofluorescence-spectral imaging for classifying soybean seed maturation stages.
• To correlate autofluorescence signals with seed physiological quality parameters.
• To develop machine learning models for accurate seed maturation stage prediction.

Main Methods:

• Soybean seeds (cv. MG/BR 46 "Conquista") were collected at five reproductive stages (R7.1 to R9).
• Autofluorescence-spectral imaging was performed using various excitation/emission combinations.
• Physical parameters, germination, vigor, and pigment dynamics were analyzed.
• Machine learning algorithms (random forest, neural network, support vector machine) were employed to classify maturation stages.

Main Results:

• Distinct autofluorescence-spectral signatures were identified for different soybean seed maturation stages.
• The excitation/emission combination of chlorophyll a (660/700 nm) and b (405/600 nm) proved effective.
• Machine learning models demonstrated high accuracy in segmenting seed maturation stages.
• Autofluorescence signals correlated with seed physiological quality indicators.

Conclusions:

• Autofluorescence-spectral imaging is a powerful, non-destructive tool for classifying soybean seed maturation.
• Chlorophyll autofluorescence wavelengths serve as reliable markers for seed maturity and physiological quality.
• This technique enables the selection of soybean seeds with superior field performance potential.