Orthogonal wavelet packet transform can divide signals (images) into multi-levels and further decompose the high-frequency part that isn't subdivided by multi-resolution analysis, thus improving the resolution of frequency and then effectively extracting the specific frequency component. This paper deduces the basic principle of wavelet packet transform, provides remote sensing image fusion algorithm based on orthogonal wavelet packet transform. Finally, an instance is given, which proves the validity and superiority of orthogonal wavelet packet transform.