Remote Sensing Image Registration Algorithm Based on Improved SURF in Wavelet Domain

Affiliation of Author(s):电子信息工程学院

Journal:Tianjin Daxue Xuebao (Ziran Kexue yu Gongcheng Jishu Ban)

Abstract:To further increase the speed of remote sensing image registration on the premise of improving registration precision, a remote sensing image registration algorithm based on improved speeded up robust features(SURF)in wavelet domain was proposed. Firstly, the reference image and the image to be registered were decomposed into the low-frequency component and high-frequency components by the wavelet transform, respectively. Then the low-frequency component served as the input image of improved SURF algorithm was proposed in this paper, thus obtaining the coarse registration point pairs. The dimension of descriptors was reduced by the principal component analysis method, and the coarse registration of feature points was achieved by the criterion of two-way registration. Next, precise registration point pairs were gradually screened out twice by using the random sample consensus(RANSAC)algorithm with different distance thresholds. Finally the parameters for geometric transformation were fitted through the least square method, and the registration image was obtained. A large number of experimental results show that the proposed method greatly improves the registration speed with higher registration accuracy compared with four algorithms, namely the scale invariant feature transform(SIFT)algorithm, the SURF algorithm, the multi-scale registration algorithm in wavelet domain using SURF, and the algorithm based on non-subsampled contourlet transform(NSCT)and SURF. © 2017, Editorial Board of Journal of Tianjin University(Science and Technology). All right reserved.

ISSN No.:0493-2137

Translation or Not:no

Date of Publication:2017-10-15

Co-author:王志来

Correspondence Author:wyq

Date of Publication:2017-10-15