To solve the problems of spectral distortion and lack of spatial details in image fusion, an image-fusion approach was proposed based on the improved guided filtering and quantum genetic algorithm.Firstly, up-sampling operation was used in multi-spectral image, and the panchromatic image was fitted by the improved guided filtering.Secondly, the new panchromatic image was optimized by using quantum genetic algorithm.Next, the multi-spectral image and panchromatic image were decomposed by wavelet transform.Then, weighted average was made to the high-frequency part, and the principle of selecting the maximum-pixel was used to the low-frequency part.Finally, the fusion image was reconstructed by adopting the inverse wavelet transformation.Experimental results show that the improved method effectively increases such image indicators as average gradient and information entropy, which can enhance the details and spectral information in image fusion, and get better visual effect.