A New Fast Multi-Source Remote Sensing Image Registration Algorithm

Article Preview

Abstract:

Aiming at the registration of multi-sensor remote sensing images, a fast and effective image registration algorithm method is presented. In this algorithm, using SIFT to extract feature points, remove mismatching points through Delaunay triangulation, and introduce distance calculation into the determination of homologous control point pairs. The effect of image registration algorithm is evaluated by subjective and objective method. Experiment results show the proposed algorithm can accurately register multi-spectral and panchromatic images with some shift, rotation angles and back-ground noises, and it can inrease the speed and precision of registration.

You might also be interested in these eBooks

Info:

Periodical:

Advanced Materials Research (Volumes 452-453)

Pages:

950-953

Citation:

Online since:

January 2012

Authors:

Export:

Price:

[1] Zhang shaohui, Shen xiaorong, Fan yaozu. A Method Based on Image Feature Extraction and Matching. Beijing University of Aeronautics Learned Journal, Vol. 34(2008), pp.516-519.

Google Scholar

[2] Liu jian, Zhang guohua, Huang linlin. Image Registration Algorithm Based on Improved SIFT. Beijing University of Aeronautics Learned Journal, Vol. 36(2010), pp.1121-1124.

Google Scholar

[3] Lowe D G. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, Vol. 60(2004), pp.91-110.

DOI: 10.1023/b:visi.0000029664.99615.94

Google Scholar

[4] Chew L P. Constrained Delaunay Triangulations. Algorithmica, Vol. 4(1989), pp.97-108.

DOI: 10.1007/bf01553881

Google Scholar