A global optimization approach for construction of panoramic mosaics

Haiying Wang; Kaihuai Qin

doi:10.1117/12.833213

30 October 2009 A global optimization approach for construction of panoramic mosaics

Haiying Wang, Kaihuai Qin

Proceedings Volume 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis; 74950F (2009) https://doi.org/10.1117/12.833213
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China

Abstract

A global optimization approach for construction of a 360° full-view panoramic mosaic is presented with global consistency in this paper. There are three primary steps in this approach. The first is a local registration using SIFT. In the second step, an efficient algorithm for finding a global consistent registration of all images is developed. The local registration matrices with the boundary condition are used for computing the initial parameters of the iterative optimization for the global consistent registration, to adjust/reduce the alignment error accumulated by the local registering and stitching of adjacent two images. Lastly, the overlapping areas of the images are stitched by the graph cut algorithm and blended with a multi-resolution method. It is proven by the experiment results in this paper that the presented approach can effectively construct panoramic mosaics with global end-to-end consistency.

Citation Download Citation

Haiying Wang and Kaihuai Qin "A global optimization approach for construction of panoramic mosaics", Proc. SPIE 7495, MIPPR 2009: Automatic Target Recognition and Image Analysis, 74950F (30 October 2009); https://doi.org/10.1117/12.833213

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