To suppress speckle noise effectively and protect edge details, a method for speckle noise suppression is proposed based on homomorphic filtering and the improved threshold integer lifting wavelet transform.Firstly, the homomrphic filtering is used for converting the multiplicative speckle noise into additive noise, and the integer lifting wavelet transform is used to decompose the image.Then, according to analysis to the shortcomings of the traditional wavelet threshold function, a new threshold function model and an adaptive threshold calculation method are designed, and threshold processing is made to the coefficients of the wavelet decomposition.Finally, inverse wavelet transform and inverse homomopgic transform are carried out and the denoised image is obtained.Simulation experiments show that the proposed algorithm can not only suppress the speckle noise effectively, but also protect the edges properly, thus it has good practicability.