Suppression of coherent noise in digital holography can obtain higher quality digital holographic reconstructed image. A suppression of holographic coherent noise method which combines bilateral filter and wavelet filter is proposed. Firstly, the holographic reconstructed image is decomposed by wavelet transformation. Then the high-frequency component in wavelet domain is suppressed by the wavelet shrinkage algorithm. The low-pass approximation component is filtered by the bilateral filter. Finally, the inverse wavelet transform is processed. Through the simulation analysis, this method improves the image peak signal-to-noise ratio 0.4 dB～2 dB, and got a higher correlation coefficientcloser with the original image. The method is verified by holographic experiment data,and the results show that this method can effectively suppress the speckle noise, and at the same time, retain more details information of the reconstructed image.