A Proposal of the Effective Recognition Method for Low-Resolution Characters from Motion Images
ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Super-Resolution of Moving Vehicles Using an Affine Motion Model
AVSS '09 Proceedings of the 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance
| Hi-index | 0.00 |
This paper proposes a new method to obtain precise projective parameters of image deformation simultaneously with non-iterative calculation by extending area-based matching and sub-pixel estimation. The method requires no "a priori" knowledge of images at all. The proposed method is based on a practical similarity model in 8-Dparameter space. Using similarity measures obtained at discrete positions in the parameter space, our method provides a highly accurate maximum position of similarity in sub-sampling resolution; that position corresponds to image deformation parameters. The estimated parameters can be used for direct multi-image super-resolution, which can directly reconstruct a high-resolution full-color image from a set of low-resolution Bayer CFA images. Experiments on the super-resolution processing were performed using real image sequences to verify the proposed method.