A Fast Orthoimage Mosaic Method for Application of Grid

Article Preview

Abstract:

Traditional orthoimage mosaic methods do not perform well in computational speed and geometric precision. This paper proposed a fast orthoimage mosaic method for the application of grid. First of all, down-sample the original images and extracts feature, and adopt SANSAC to estimate the relative initial homography; second, refine homography matrix by Levenberg-Marquardt method and use the sparse bundle adjustment method to estimate the precise homography matrix; Third, passed the homography matrix to the original level of image by the homography relationship of the down-sampling and original image. Finally, synthesize the mosaic image. Our experiments showed that the method combined the down-sampling homography relationship of original image with sparse bundle adjustment organically, which effectively improved the speed and obtained geometric seamless mosaic.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

685-689

Citation:

Online since:

January 2012

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2012 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

[1] Bay H, Tuytelaars T, Van Gool L. SURF: Speeded up robust features [A]. In: European Conference on Computer Vision[C], 2006, 1: 404-417.

DOI: 10.1007/11744023_32

Google Scholar

[2] Bo Wang. A fast UAV image mosaic method generating geometric consistency [J]. Geometrics Science, (2011).

Google Scholar

[3] Lourakis M I A, Argyros A A. SBA: A software package for generic sparse bundle adjustment [J]. ACM Transactions on Mathematical Software, 2009, 36(1): 1-30.

DOI: 10.1145/1486525.1486527

Google Scholar

[4] Bo Wang. Automatic Mosaic of UAV Remote Sensing Image Based on Features [D]. Thesis for master degree, Information and Engineering University, 2011. 06.

Google Scholar

[5] Richard H, Andrew Z. Computer Vision in multi-view geometry [M]. Sui Wei translates. Anhui: Anhui University Press, 2002:52-82.

Google Scholar

[6] Brown, M. and Lowe, D. Automatic panoramic image stitching using invariant features [J]. International Journal of Computer Vision, 2007, 74(1): 59–73.

DOI: 10.1007/s11263-006-0002-3

Google Scholar