An asynchronous near-infrared correlation cut-spectrum detection method to reduce the influence of brands on the discrimination model of urea-doped milk

Abstract

The invention belongs to the field of detection methods, and relates to the detection of adulterated milk, especially an asynchronous near-infrared correlation cut-spectrum detection method that reduces the influence of brands on the discrimination model of urea-doped milk. The information shared by the system, that is, the inherent component information of adulterant urea and different brands of milk enters the model, thereby reducing the impact of brands on the discrimination model and improving the accuracy of discrimination; the present invention utilizes asynchronous near-infrared correlation Modeling and analysis of the cutting spectrum at the location can not only reduce the influence of milk brand on the model, but also improve the correct rate of discrimination, and compared with the entire asynchronous two-dimensional correlation spectrum matrix, the characteristic data is less and the modeling efficiency is high; the method used in the present invention It is simple, scientific, and has high analysis efficiency and discrimination accuracy, and can be extended to other food adulteration detection.

Description

technical field [0001] The invention belongs to the field of detection methods, and relates to the detection of adulterated milk, in particular to an asynchronous near-infrared correlation cut-spectrum detection method that reduces the influence of brands on the discrimination model of urea-doped milk. Background technique [0002] Milk is the only food in the initial stage of people's life. With the rapid development of society and the improvement of people's living standards, milk has increasingly become a daily drink in our daily life. Milk is very rich in trace elements. Milk has the functions of moisturizing the intestines and dryness, and also has the function of delaying facial aging and slowing down the aging of people's brain, so it is welcomed by consumers. However, many unscrupulous traders, in order to obtain huge profits, often add some adulterants different from the inherent components of milk to milk. These adulterated milk directly endanger human health, and...

AI Technical Summary

Problems solved by technology

Therefore, in view of the deficiencies in the prior art, the present invention provides a detection method based on asynchronous near-infrared correlation cut-spectrum shared feature information to reduce the influence of milk brands on the adulterated milk discrimination model, starting from the extraction of characteristic near-infrared spectrum information, How to effectively extract only the change characteristic information caused by adulterated components, while ignoring the change information caused by sample differences, and only let the change information caused by adulterated components enter the model (for unknown samples, use the same method Extraction), so as to reduce the impact of sample differences on the model due to factors such as different brands, origins, production batches, etc., this method has not been reported in relevant literature at home and abroad

Images