Use of near infrared hyperspectral imaging as a nondestructive method of determining and classifying shelf life of cakes

Abstract

Many types of cake deteriorated rapidly due to microbial infection, which gives them a short shelf life. Accordingly, near-infrared hyperspectral imaging (NIR-HSI), in the range of range of 935–1720 nm, was tested to determinate whether it could be used as a nondestructive method to determine the shelf life and classify cakes based on microorganism infections during storage. The average spectrum from a region of interest (ROI) in the spectral image of each sample was acquired by NIR-HSI. Partial least squares regression (PLSR) was used to establish the model in order to predict storage time of sponge cakes. The model proved accurate with a correlation coefficient (R) of 0.835 and the root mean square error of prediction (RMSEP) of 1.242 days. Partial least squares discriminant analysis (PLS-DA) was applied to establish the classification model for distinguishing between non-expired and expired of sponge cakes. The results showed the accuracy of prediction was 91.3%. The predictive images showed different colors based on their storage time that could be inspected visually. Therefore NIR-HSI was shown to have potential to be used for predicting storage time of cakes and classifying cakes into expired and non-expired, which has potential for application in the bakery industry.