Oblique aerial image matching based on iterative simulation and homography evaluation

作者：

Highlights：

• Iterative transform simulation is proposed to reduce the computational complexity of image matching.

• A geometric method to evaluate the homography matrix is proposed to filter out wrong matching results.

• Superior performance in terms of speed while maintaining accuracy in matching oblique aerial images.

摘要

•Iterative transform simulation is proposed to reduce the computational complexity of image matching.•A geometric method to evaluate the homography matrix is proposed to filter out wrong matching results.•Superior performance in terms of speed while maintaining accuracy in matching oblique aerial images.

论文关键词：SIFT descriptor,Feature matching,Aerial image,Viewpoint change,Homography matrix

论文评审过程：Received 29 December 2017, Revised 20 October 2018, Accepted 21 October 2018, Available online 26 October 2018, Version of Record 1 November 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.10.027