Coarse-to-fine geometric and photometric image registration

Jieping Xu; Jin Liu; Zongfu Huang; Yonghui Liang

doi:10.1117/12.2281123

24 October 2017 Coarse-to-fine geometric and photometric image registration

Jieping Xu, Jin Liu, Zongfu Huang, Yonghui Liang

Proceedings Volume 10462, AOPC 2017: Optical Sensing and Imaging Technology and Applications; 1046202 (2017) https://doi.org/10.1117/12.2281123
Event: Applied Optics and Photonics China (AOPC2017), 2017, Beijing, China

Abstract

This paper presents a technique that performs coarse-to-fine image registration both in spatial and range domain. The goal of image registration is to estimate geometric and photometric parameters via minimization of an objective function in the least square sense. In order to reduce the probability of falling into a local optimal solution, the algorithm employs a coarse-to-fine strategy. In the coarse step, an illumination offset and contrast invariant feature detector which is named SURF is used to estimate affine motion parameters between the reference image and the target image, and then the intensity of corresponding pixels is used to directly estimate contrast and bias parameters based on RANSAC. In the fine step, the estimated parameters obtained in the coarse step are used as a good initial estimation, and photometric and affine motion parameters are refined alternatively via minimizing the objective function. Experiments on simulated and real images show that the proposed image registration method is superior to the feature-based method used in the coarse step and the groupwise image registration algorithm proposed by Bartoli.

Citation Download Citation

Jieping Xu, Jin Liu, Zongfu Huang, and Yonghui Liang "Coarse-to-fine geometric and photometric image registration", Proc. SPIE 10462, AOPC 2017: Optical Sensing and Imaging Technology and Applications, 1046202 (24 October 2017); https://doi.org/10.1117/12.2281123

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available