Non-destructive prediction method of Yuan'an yellow tea boredom time based on least squares support vector machine

Abstract

The non-destructive prediction method of the yellowing time of Yuanan yellow tea based on the least squares support vector machine, including: collection and classification of fresh leaf samples; scanning to obtain the near-infrared spectra of fresh leaf samples with different yellowing time; preprocessing and eliminating the sample spectrum After the noise information, the sample spectrum is converted into pairs of data points; then all the spectral data are divided into 20 sub-intervals, and the least squares support vector machine method model of each sub-interval data is established to screen out the best model. Subinterval data; use principal component analysis to extract and compress the best spectral subinterval information; use the principal component score as the input value, continuously adjust the number of neurons and transfer function, and establish a general regression structure artificial neural network prediction model; the model is robust sex test. The rapid, accurate and objective prediction of the yellowing time of the yellow tea samples is realized, and the purpose of improving the accuracy of predicting the yellowing time and enhancing the practicability of the model is achieved.

Description

technical field [0001] The invention relates to a method for predicting the time of dull yellow tea, and more specifically relates to a nondestructive prediction method for the time of dull yellow tea of ​​Yuan'an yellow tea based on a least squares support vector machine. Background technique [0002] Yellow tea is one of the six major teas in my country, and Yuan'an yellow tea has the longest history in Hubei Province. Hubei Yuan'an has a mild climate, abundant rainfall, loose and fertile soil, and a good ecological environment is very beneficial to the growth of tea trees, which ensures the high quality of tea. Therefore, Yuan'an Yellow Tea is known as the best tea in Hubei, and it has always been a relatively popular tea in the tea market. The tea products are in short supply every year and are deeply loved by everyone. [0003] During processing, the picking standards for fresh leaves of Yuanan yellow tea are generally single bud, one bud and one leaf, and one bud and ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the defects of the existing Yuanan yellow tea, such as manual recording of the dull yellowing time, which cannot accurately grasp the dull yellowing time in real time, and easily lead to a decrease in the quality of the tea, and provide Yuanan yellow tea dull yellow tea based on the least squares support vector machine. A Lossless Prediction Method of Time

Images