Non-Destructive Detection of the Freshness of Air-Modified Mutton Based on Near-Infrared Spectroscopy

Area of Science:

• Food Science
• Analytical Chemistry
• Spectroscopy

Background:

• Meat freshness monitoring is crucial for food safety.
• Traditional lamb meat quality assessment methods are invasive and cumbersome.
• Modified atmosphere packaging (MAP) is widely used for meat preservation.

Purpose of the Study:

• To develop a rapid, non-destructive method for assessing the freshness of MAP lamb meat.
• To evaluate the effectiveness of near-infrared (NIR) spectroscopy combined with multi-parameter fusion for quality assessment.
• To compare different spectral preprocessing and feature selection methods for model optimization.

Main Methods:

• Near-infrared spectroscopy was used to collect spectral data from lamb meat under various conditions (temperature, MAP).
• Physicochemical indicators (rebound, TVB-N) were measured and correlated with spectral data.
• Various spectral preprocessing techniques (MSC, SNV, SG-2nd) and feature selection algorithms (SPA, UVE) were employed.
• Partial Least Squares (PLS) and Support Vector Machine (SVM) models were constructed to predict meat quality parameters.

Main Results:

• The SG-2nd-SPA-PLSR model achieved high accuracy (R²p = 0.94) for predicting rebound values.
• The MSC-UVE-SVM model demonstrated excellent accuracy (R²p = 0.95) for predicting TVB-N content.
• NIR spectroscopy successfully correlated with key meat quality indicators under different MAP conditions.

Conclusions:

• NIR spectroscopy provides a viable, non-destructive tool for rapid assessment of lamb meat freshness and quality.
• The developed models offer a theoretical and technical foundation for portable devices for real-time meat quality monitoring.
• This approach enhances food safety by enabling quick and accurate evaluation of MAP lamb meat quality.