This paper comprehensive analyzed characteristics of Redundant Wavelet Transform, put forward a multi-focus image fusion algorithm based on Redundant Wavelet Transform. That researched the selection principles about low and high frequency coefficients according to different frequency domain after Redundant Wavelet Transform. In choosing the low frequency coefficients, the concept of local area variance was chose to measure definition. In choosing the high frequency coefficients, according to the characteristics of regional strength variation, the concept of choosing the biggest absolute value was adopted. Finally, the proposed algorithm in this article was applied to experiments of multi-focus image fusion. According to simulation results, the proposed algorithm hold useful information from source multiple images quite well, improved visual effects of fused image.