Abstract: In this paper, a method based on wavelet transform, which is used to analyze near-infrared spectrum and abstract useful information, is discussed on the case of reducing sugar of vinegar. The results show that the spectrum transform can reflect their correlation with the element quantitative. By the comparison with the multiple regression analysis for the vinegar absorbed spectrum, the following remarks are found out: the wavelet coefficient 7-scale transformed has the most intimated correlation with the reducing sugar. The prediction difference uncertainty (which is 0.211) of ternary regression analysis is the least, which is 0.058 lower than that of primary spectrum. The method of wavelet transform improved the quantitatively analyzing accuracy, therefore, it is as a reference, which is used to develop the application of near-infrared analyzing technology in vinegar on-line measurement.