Fast software image stabilization with color registration

We present the formulation and implementation of an image stabilization system capable of stabilizing video with very large displacements between frames. A coarse-to-fine technique is applied in resolution and in model spaces. The registration algorithm uses phase correlation to obtain an initial estimate for translation between images; then Levenberg-Marquardt method for nonlinear optimization is applied to refine the solution. Registration is performed in color space, using a subset of the pixels selected by a gradient-based sub-sampling criterion. This software implementation runs at 5 Hz on non-dedicated hardware (Silicon Graphics R10000 workstation).

[1]  C. D. Kuglin,et al.  The phase correlation image alignment method , 1975 .

[2]  Azriel Rosenfeld,et al.  Multiresolution image processing and analysis , 1984 .

[3]  Lisa M. Brown,et al.  A survey of image registration techniques , 1992, CSUR.

[4]  Michal Irani,et al.  Recovery of ego-motion using image stabilization , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[5]  Kristin J. Dana,et al.  Real-time scene stabilization and mosaic construction , 1994, Proceedings of 1994 IEEE Workshop on Applications of Computer Vision.

[6]  Richard Szeliski,et al.  Image mosaicing for tele-reality applications , 1994, Proceedings of 1994 IEEE Workshop on Applications of Computer Vision.

[7]  Rama Chellappa,et al.  Automatic Digital Image Stabilization , 1996 .

[8]  Rama Chellappa,et al.  Fast electronic digital image stabilization , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[9]  Eric Krotkov,et al.  Automatic Mountain Detection and Pose Estimation for Teleoperation of Lunar Rovers , 1997, ISER.

[10]  Fabio Gagliardi Cozman,et al.  Automatic mountain detection and pose estimation for teleoperation of lunar rovers , 1997, Proceedings of International Conference on Robotics and Automation.