Lotus root starch adulteration identification method based on machine learning

Abstract

The invention relates to a lotus root starch adulteration identification method based on machine learning, which comprises the following steps of purchasing a proper amount of fresh lotus roots on the internet to prepare pure lotus root starch, preparing an adulterated lotus root starch sample for use on a machine, collecting spectral data of lotus root starch samples with different doping proportions, based on the obtained spectral data, establishing a machine learning clustering model for prediction, and performing adulteration prediction on the lotus root starch sample to be detected based on the established clustering model. According to the method, near infrared spectrum data of lotus root starch samples with different doping proportions are collected as an original data set, so that a machine learning clustering model is established, and the model can effectively identify adulterated lotus root starch samples. The method is simple and rapid in detection, the detection efficiency can be remarkably improved, a new method is provided for simplifying lotus root starch quality identification, and the method has very high practicability and wide application prospects.

Description

technical field [0001] The invention relates to the field of food detection, in particular to a method for identifying adulteration of lotus root starch based on machine learning. Background technique [0002] Lotus root starch is a traditional health food in my country. Lotus root starch and its related products are deeply loved by consumers at home and abroad and paid attention to by food researchers. Lotus root contains dopa, catechin, gallic acid, and catechin, which are natural antioxidants with development value. With the continuous promotion of edible lotus root powder and the continuous expansion of market share, the quality of lotus root powder is mainly faced with the following problems: Driven by profit, there are potential adulteration problems in the market, such as posing fake products as real products, shoddy products as good ones, or substituting them with other cheap similar products , such as mixing cassava flour, sweet potato flour and corn flour into hig...

AI Technical Summary

Problems solved by technology

The traditional detection technology of lotus root starch has its limitations. The method of identification through appearance has the disadvantages of strong subjectivity and large errors; the use of national standard optical microscopes and polarizing microscopes can only identify starch granules similar to typical lotus root starch, but it is impossible to identify the starch granules of lotus root starch. It can be distinguished from other starch granules similar to lotus root starch, such as starch granules such as potatoes; the scanning electron microscope method cannot identify cassava flour and other atypical small lotus root starch particles that are very similar to the ultrafine morphology of lotus root starch; differential scanning calorimetry is for experimental conditions The selection requirements are high, and the selection of different experimental conditions has a great impact on the results. Therefore, the experimenters must have sufficient experience, and there is a shortage in the breadth of application.

Images