Image fusion is merged to form a single image, which is more informative than the single input image in quality and appearance. In order to represent the source images effectively and completely, a novel guided image filtering based fusion (GFF) for multi-focus images is proposed. Guided image filtering decomposes an image into a base layer and a detail layer. The base layer contains large scale variations in intensity and the detail layer captures small scale details. The focus region of the base layer is detected according to energy of image gradient (EOG). The focus region of the detail layer is detected using spatial frequency (SF). The base and the detail layer are superposed to construct the final fused image. Experimental results demonstrate that the proposed method can obtain state-of-the-art performance for fusion of multi-focus images.