Image fusion has become an important and powerful technique for image analysis and computer vision. This paper presents a novel multiresolution image fusion method, which is based on wavelet transform combing with an effective fusion scheme. The main contribution of this research is that by considering the physical meaning of the wavelet coefficients, a selection scheme that treats the coefficients in different ways is proposed. This scheme selects the coefficients in the high frequency bands by a wavelet entropy based strategy, while selects the coefficients in the low frequency band by a variance based strategy. The performance of the proposed fusion method is compared with several existing fusion techniques. Comparison results show that the proposed method can effectively fuse the images with less error.