In the view of the situation that single-sensor image cannot fully reflect the scene information efficiently. A fusion method of infrared and visible images based on discrete wavelet transform is presented and comparatively analyzed with traditional methods. Firstly, the wavelet multi-scale decomposition technique is applied to the source images that will be fused to give a series of sub-band coefficient. Feature extraction and weighted average with adaptive weighting factors are used to process the high-frequency coefficients. A strategy of the absolute value comparing is adopted to the low-frequency coefficients. Finally, the fusion image is reconstructed by multi-scale wavelet inversing transformation for low frequency and high frequency coefficients. Experimental results demonstrate that infrared and visible images can be more effectively fused by the algorithm presented than traditional methods.