Image mosaicing by stratified matching

作者:

Highlights:

摘要

We present a robust method for automatically matching points over two images for image mosaicing: after extracting feature points using a feature detector, we progressively estimate the rotation, the scale change, and the projective distortion between the two images by random voting and variable template matching. Using real images, we demonstrate that our method allows accurate image mosaicing even when conventional methods fail.

论文关键词:Template matching,Feature point correspondence,Robust estimation,LMedS,Image mosaicing,Homography

论文评审过程:Received 23 September 2002, Revised 27 June 2003, Accepted 2 July 2003, Available online 19 September 2003.

论文官网地址:https://doi.org/10.1016/j.imavis.2003.07.001