Aflatoxin contaminated degree detection by hyperspectral data using band index

Area of Science:

• Agricultural Science
• Food Science
• Analytical Chemistry

Background:

• Aflatoxins are highly toxic and carcinogenic compounds found in agricultural products.
• Accurate detection of aflatoxin contamination is crucial for food safety and public health.
• Current detection methods can be time-consuming and may not be suitable for real-time monitoring.

Purpose of the Study:

• To investigate the feasibility of using hyperspectral imaging technology for detecting aflatoxin contamination in peanuts.
• To develop a machine learning-based method for quantifying aflatoxin levels.
• To evaluate the effectiveness of proposed fluorescence indexes and support vector machine (SVM) models.

Main Methods:

• Preparation of peanut kernels with varying aflatoxin concentrations.
• Acquisition of hyperspectral images (400-720 nm) under UV light (365 nm).
• Development and application of fluorescence indexes (RI, DRI, RRI, NDRI) and a Radial Basis Function Support Vector Machine (RBF-SVM) model for detection and regression analysis.

Main Results:

• The Difference Radiation Index (DRI) demonstrated optimal performance among the tested indexes.
• The SVM model achieved a 5-fold cross-validation accuracy rate of 95.5%.
• The model yielded a mean square error (MSE) of 0.0223 and a correlation coefficient (R) of 0.9785, indicating high accuracy in regression analysis.

Conclusions:

• Hyperspectral imaging combined with machine learning, particularly using the DRI index and SVM, is a feasible and accurate method for detecting aflatoxin contamination in peanuts.
• The developed method shows significant potential for online, real-time detection and grading of aflatoxin in agricultural products.
• This approach contributes positively to enhancing food safety standards and quality control in the agricultural industry.