Abstract: Spatial heterodyne spectroscopy (SHS) is a new spectroscopic technique which applies to astrometry and atmospheric remote sensing. A method of using wavelet transform to correct the baseline was proposed to eliminate the baseline drift during its application. Through the analysis of the influences of the different wavelet functions and decomposition layers on the results of baseline correction, the original spectrum was decomposed by wavelet transforms to get detail coefficients and approximate coefficients, and set the approximate coefficients to zero, the spectrum was reconstituted to realize the baseline correction. In the end, the results were compared with the results of the threshold fitting correction. The results show that the correlation coefficient of the two methods is 0.999 9 and the results are consistent, but the method of wavelet transform is more than ten times the time of the procedure. It is proved that using wavelet transform to carry out the baseline correction of spatial heterodyne interferometry is an effective, time saving and convenient method.