Spectrum identification method for high-temperature sterilized milk, pasteurized milk and pasteurized milk doped with high-temperature sterilized milk

Abstract

The invention belongs to the technical field of dairy product analysis, and particularly relates to a spectrum identification method for high-temperature sterilized milk, pasteurized milk and pasteurized milk doped with the high-temperature sterilized milk. The method comprises the steps of 1) taking pasteurized milk and high-temperature sterilized milk, and simulating different adding proportions of the pasteurized milk to prepare samples; 2) scanning the samples in a medium infrared spectrum range to obtain medium infrared spectrum data; 3) preprocessing the original mid-infrared spectrum, and removing abnormal values; 4) dividing the preprocessed data set into a training set and a test set according to stratified sampling; 5) screening a modeling spectral band; 6) detecting the sample identification model on the training set through 10-fold cross validation by using a nearest neighbor algorithm and the like, and evaluating and screening the model by using accuracy and kappa coefficients; and 7) verifying the optimal model. According to the method, less wave points are used for modeling, the operation cost is reduced, and the identification rate and accuracy are improved.

Description

technical field [0001] The invention belongs to the technical field of milk product analysis, in particular to a rapid identification method for pasteurized milk mixed with high-temperature sterilized milk, pasteurized milk and high-temperature sterilized milk Background technique [0002] Milk contains protein, fat, carbohydrates, vitamins, minerals and essential amino acids and other nutrients, but freshly squeezed raw milk also contains a variety of microorganisms, so it is necessary to destroy or reduce pathogenic microorganisms in milk and reduce enzymes To ensure milk safety and prolong shelf life, raw milk is usually heat-treated (Liu, Grosvenor et al. 2019). The most commonly used heat treatment methods are pasteurization (70-85°C for 15-20 seconds) and high temperature sterilization (UHT: 135-150°C for 1-10 seconds). Studies have shown that heat treatment will reduce the antioxidant capacity of milk. Compared with raw milk and pasteurized milk, high-temperature ste...

AI Technical Summary

Problems solved by technology

The data output by the mid-infrared spectrometer is a matrix of n×1060 (n is the sample size), the data is huge, and it is difficult to avoid data incompleteness, inconsistency, and is easily disturbed by noise (errors or outliers). Low-quality data will lead to Poor data mining results, so some method is needed to preprocess the output data

Images