An approach based on contourlet image fusion and reformulated fuzzy clustering for change detection in satellite images is introduced in this paper. In this approach fusion of images is used to produce difference image from log ratio and mean ratio images. An optimal difference image should retain the unchanged areas and show the changed areas. So contourlet image fusion is used to generate the difference image. Processing the difference image means to discriminate changed regions from unchanged regions using reformulated fuzzy local information c means algorithm. It is used because it is very sensitive to noise. Experimental result shows that this approach provides better performance than the previous methods.