Youngwook Paul Kwon

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2014-121

May 20, 2014

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-121.pdf

We propose a new segment-based registration system for aerial images of the same scene taken at different times, from different view points, and/or by different sensors. We introduce a quantitative characterization of the registration difficulty for a given pair of images. Targeting high registration difficulty input, we exploit on linear edges in images. In the first step of our registration process, we detect line segments in each image. Next we conduct a merging step on the detected line segments. Finally, using the merged line segments as input, we generate possible hypothesis transformations by choosing three segments in each image. Our collinearity score metric for the transformations balances considerations of angular and perpendicular distances. After scoring each hypothesis transformation, the highest-scoring one is selected. For high registration difficulty image pairs, our algorithm shows significant improvement compared to publicly accessible image registration codes.

Advisors: Alexei (Alyosha) Efros

---

BibTeX citation:

@mastersthesis{Kwon:EECS-2014-121,
    Author= {Kwon, Youngwook Paul},
    Title= {Line Segment-based Aerial Image Registration},
    School= {EECS Department, University of California, Berkeley},
    Year= {2014},
    Month= {May},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-121.html},
    Number= {UCB/EECS-2014-121},
    Abstract= {We propose a new segment-based registration system for aerial images of the same scene taken at different times, from different view points, and/or by different sensors. We introduce a quantitative characterization of the registration difficulty for a given pair of images. Targeting high registration difficulty input, we exploit on linear edges in images. In the first step of our registration process, we detect line segments in each image. Next we conduct a merging step on the detected line segments. Finally, using the merged line segments as input, we generate possible hypothesis transformations by choosing three segments in each image. Our collinearity score metric for the transformations balances considerations of angular and perpendicular distances. After scoring each hypothesis transformation, the highest-scoring one is selected. For high registration difficulty image pairs, our algorithm shows significant improvement compared to publicly accessible image registration codes.},
}

EndNote citation:

%0 Thesis
%A Kwon, Youngwook Paul 
%T Line Segment-based Aerial Image Registration
%I EECS Department, University of California, Berkeley
%D 2014
%8 May 20
%@ UCB/EECS-2014-121
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-121.html
%F Kwon:EECS-2014-121