Speed Up Robust Features（SURF） is an improvement of Scale Invariant Feature Transform（SIFT）. When it is used to match remote sensing images

SURF can significantly increase matching speed

but slightly decrease matching accuracy. Thus

a r emote sensing image matching method based on Non-Subsampled Contourlet Transform（NSCT） and SURF is proposed in this paper. First

the remote sensing image to be matched and the reference image are decomposed by NSCT. Two corresponding l ow-frequency images are obtained. Then

to reduce the influence of high-frequency noise on matching results

two low-frequency images are inputted to the SURF algorithm to obtain pre-matching results. Finally

to solve the error matching problem of the SURF algorithm

the parameters of the transform model are solved by pre-matching results

and the mismatching is eliminated by using the random sample consensus algorithm. A large number of experiments were conducted

and the results show that compared with the SIFT and SURF algorithms

the proposed algorithm improves the matching speed

as well as the matching accuracy

and exhibits good performance in terms of resisting rotation

noise

and brightness changes.