Multiwavelets is a new development to the body of wavelet theory. Multiwavelets simultaneouslly offers orthogonality, symmetry, and short support, which is not possible with scalar two-channel wavelet systems. A new theory and algorithm for speckle reduction of ultrasound (US) image with multiwavelets multiple resolution analysis (MRA) are presented and investigated in this paper. The decomposition ratio and reconstruction error of two kinds of multiwavelets are evaluated and the optimum pre-filters suitable for the two kinds of multiwavelets are obtained. Fully and clearly analytic expression of multiwavelet transformation is given .The experiment of speckle reduction to US image is implemented by choosing local error as threshold. Results show that multiwavelet transformation is useful for speckle reduction and there have little noise and reserve important image features such as boundary when compared to results obtained from existing the denoising methods alone.