• Opto-Electronic Engineering
  • Vol. 39, Issue 8, 26 (2012)
RUI Ting1、*, ZHANG Sheng-ao1, ZHOU You2, SUN Zheng1, and CAO Peng1
Author Affiliations
  • 1[in Chinese]
  • 2[in Chinese]
  • show less
    DOI: 10.3969/j.issn.1003-501x.2012.08.004 Cite this Article
    RUI Ting, ZHANG Sheng-ao, ZHOU You, SUN Zheng, CAO Peng. Registration of Multi-sensor Images Based on Harris Corner with SIFT Description[J]. Opto-Electronic Engineering, 2012, 39(8): 26 Copy Citation Text show less

    Abstract

    Multi-sensor image registration is very difficult because of their different imaging principles. Referring to the registration of infrared and visible images, a registration algorithm is proposed based on the Harris corner with SIFT description. First, the scale spaces were built, and the scale-invariant Harris corners were selected as feature points. Then, by improving SIFT description method of feature points, ring operator was used to describe feature points. Finally, feature points were matched by the two-way nearest neighbor algorithm, and least squares method was employed to find optimal solutions to affine transform equations. Experimental results show that the registration algorithm is accurate, efficient and stable, and has a good registration result.
    RUI Ting, ZHANG Sheng-ao, ZHOU You, SUN Zheng, CAO Peng. Registration of Multi-sensor Images Based on Harris Corner with SIFT Description[J]. Opto-Electronic Engineering, 2012, 39(8): 26
    Download Citation