The present invention provides a micro seismic data denoising method based on wavelet transformation. The method comprises the concrete steps of (1) carrying out the wavelet decomposition on the micro seismic data to obtain the finest scale of high frequency and low frequency wavelet coefficients; (2) ranking each scale of high frequency and low frequency wavelet coefficients again from large to small; (3) estimating the noise variance of the high frequency and low frequency wavelet coefficients at different scales, and calculating the mid-value and the minimum value in the new sequence; (4) calculating the threshold value of each scale of high frequency and low frequency wavelet coefficients according to constraint factors; (5) retaining various scales of coefficients greater than the absolute values of the threshold values, and calculating the new threshold values according to the step (4) for the coefficients less than the absolute values of the threshold values; (6) repeating the steps (2) to (5), and saving various scales of new high frequency and low frequency wavelet coefficients after processing; (7) obtaining the micro seismic data after denoising by the wavelet inverse transformation. The micro seismic data denoising method based on wavelet transformation of the present invention solves the problem that the micro seismic data is high in signal to noise ratio, and has the waveform distortion after denoising.