In recent years, data collected from remote sensing satellite and aerophotography have been showing a geometric sequence increase. A method of Scale Invariant Feature Transform (SIFT) algorithm could be employed for the automatic geometric fine correction. This method could avoid the impact of the rotation and zooming of template matching during the image matching process, and it can also save the labor during the image processing operation. Based on the SIFT algorithm, this paper proposes a two-step method, which firstly conducts coarse match on feature points, and then further conducts fine correction on the coarsely matched feature points by using the least squares technique. The result indicates that, this method is an effective automatic matching method for remote sensing images.