论文信息 - A new approach to corner extraction and matching for automated image registration

A new approach to corner extraction and matching for automated image registration

Automatic image registration is an important process in remote sensing applications. In this paper, an efficient automatic system is resented for the estimation of the similarity transform parameters from partially overlapping images. Several aspects of the proposed system including corner extraction, corner matching and iterative transform parameter estimation are discussed in details. Experimental results using remote sensing images are presented.

Farzin Deravi | Salem Alkaabi | F. Deravi | Salem Alkaabi

[1] Farzin Deravi,et al. Block matching in rotated images , 2005 .

[2] Siamak Khorram,et al. A feature-based image registration algorithm using improved chain-code representation combined with invariant moments , 1999, IEEE Trans. Geosci. Remote. Sens..

[3] Ian Dowman,et al. An improved model for automatic feature-based registration of SAR and SPOT images , 2001 .

[4] Farzin Deravi,et al. Candidate pruning for fast corner detection , 2004 .

[5] Hui Henry Li,et al. Automatic visual/IR image registration , 1996 .

[6] M. Roederer,et al. Probability binning comparison: a metric for quantitating univariate distribution differences. , 2001, Cytometry.

[7] K. Mardia,et al. A review of image-warping methods , 1998 .

[8] Farzin Deravi,et al. Gradient direction similarity measure , 2003 .

[9] Carlo Tomasi,et al. Good features to track , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[10] Rama Chellappa,et al. A computational vision approach to image registration , 1993, IEEE Trans. Image Process..