Rapid and nondestructive detection of sorghum adulteration using optimization algorithms and hyperspectral imaging

Area of Science:

• Agricultural Science
• Analytical Chemistry
• Food Science

Background:

• Sorghum adulteration poses a significant challenge to grain quality and trade.
• Accurate and rapid detection methods are crucial for maintaining sorghum integrity.

Purpose of the Study:

• To develop and validate a novel model for detecting adulteration in sorghum using hyperspectral imaging (HSI).
• To establish a rapid, non-destructive method for assessing sorghum grain purity and variety identification.

Main Methods:

• Hyperspectral imaging technology (HSI) for data acquisition.
• Watershed algorithm for hyperspectral data extraction from sorghum grains.
• Principal Component Analysis (PCA) and Clustering Analysis (CA) for sample cleaning.
• Partial Least Squares Discriminant Analysis (PLS-DA) for variety identification and adulteration ratio determination.

Main Results:

• The model achieved high accuracy in identifying sorghum varieties (96% for validation set) and detecting adulteration (91% for adulterated samples).
• Comprehensive model accuracy exceeded 90%, demonstrating its effectiveness.
• A sorghum distribution map with color-coded varieties was generated, facilitating visual assessment.

Conclusions:

• This research presents the first application of HSI combined with PCA and CA for sorghum adulteration identification.
• The developed model offers a rapid, non-destructive, and accurate solution for detecting adulteration in sorghum grains.
• The findings support the potential of HSI technology in agricultural quality control and food safety applications.