Abstract: A new method of model construction based on continuous wavelet transform (CWT) and independent component regression (ICR) was proposed.In its application to near infrared spectrometry,the data of IR spectra were treated with CWT to remove the noise and spectral background,their characteristic matrix were then extracted by the independent component analysis,and finally,determination of the individual components of the sample was performed by the regression analysis.This new chemometric method has been applied to the simultaneous determination of contents of moisture,fat and protein in meat samples,showing that the results predicted by CWT-ICR method were in consistency with the results obtained by chemical analysis.