Watermelon sugar degree rapid modeling method based on secondary spectrum recombination

Abstract

The invention discloses a watermelon sugar degree rapid modeling method based on secondary spectrum recombination. The method comprises the following steps: collecting diffuse transmission spectrums of a watermelon, and selecting spectrums of J wavelengths within a high-signal-noise-ratio spectrum range; measuring sugar degree values of the watermelon; adopting a uniformly-spaced extraction spectrum method or a uniformly-spaced average spectrum method to reconstruct new spectrum matrixes, arranging uniformly-spaced points j, enabling j to be no smaller than 1 and no larger than J / 2 and be a positive integer, reconstructing the spectrums of J wavelengths, making comparisons of modeling effects of the new spectrum matrixes, and performing self-adaption optimization of the new spectrum matrixes with the correlation coefficient to be over 0.8; performing spectrum pretreatment on the new spectrum matrixes with the correlation coefficient to be over 0.8; and using diffuse transmission spectrums screened out and the sugar degree values of the watermelon to build a prediction model of the watermelon sugar degree, and achieving on-line detection of the watermelon sugar degree. The watermelon sugar degree rapid modeling method can also be used for detection research of internal quality of other fruits such as sugar degree, acidity and hardness.

Description

technical field [0001] The invention relates to a watermelon sugar content modeling method, in particular to a watermelon sugar content rapid modeling method based on secondary spectral recombination. Background technique [0002] With the development of science and technology and the continuous improvement of people's living standards, people have higher and higher requirements for the appearance, flavor and nutrition of fruits. Traditional fruit internal quality inspection methods are all destructive testing, which cannot meet the development needs of nondestructive inspection and real-time grading of fruit internal quality. [0003] With the development of non-destructive testing technology based on spectral analysis for many years, the price of the instrument has become civilians. Even because this method does not need to use chemical reagents, the detection cost is lower than some traditional national standard methods, and its operation is simple, and it can realize fas...

Images