[The pear acidity quantified analysis using PLS methods and Fourier transform near-infrared spectroscopy]

Area of Science:

• Agricultural Science
• Analytical Chemistry
• Spectroscopy

Context:

• Assessing fruit quality is crucial for the agricultural industry.
• Non-destructive methods are increasingly preferred for quality control.
• Fourier transform near-infrared (FT-NIR) spectroscopy offers a rapid and non-invasive analytical technique.

Purpose:

• To develop and validate a Fourier transform near-infrared (FT-NIR) spectroscopy method for predicting the quality of intact Xueqing pears.
• To establish a robust partial least squares (PLS) regression model for correlating spectral data with laboratory measurements.
• To identify optimal spectral regions and model parameters for accurate quality prediction.

Summary:

• FT-NIR spectra of intact Xueqing pears were acquired using diffuse reflectance.
• Partial least squares (PLS) regression was employed to model the relationship between spectral data and laboratory quality parameters.
• The optimal model, utilizing the 5452-12285 cm(-1) wavenumber range and 7 factors, achieved a correlation coefficient of 0.79 and low prediction errors (0.019).

Impact:

• Demonstrates the feasibility of using FT-NIR spectroscopy for non-destructive quality assessment of Xueqing pears.
• Provides a validated method for precise quality prediction, potentially reducing reliance on traditional destructive testing.
• Highlights the effectiveness of strategic wavelength selection in enhancing the accuracy of spectroscopic models for agricultural products.