For wavelet-based image compression algorithm, it focuses on the hierarchical trees algorithm (SPIHT) and makes description of the principle and process of the algorithm. Apply MATLAB to convert the gray image by wavelet transformation, then conduct coding on wavelet transform image coefficients SPIHT. The received bit stream obtained by decoding and inverse wavelet transform the reconstructed image simulation. The results have been reconstructed image signal to noise ratio and the operation time. Experimental results show that the algorithm can quickly and efficiently encode the image processing.