Blind Superresolution with Generalized Cross-Validation using Gauss-Type Quadrature Rules

| Superresolution reconstruction produces a high resolution image from a set of aliased low resolution images. We model the low resolution frames as blurred and down-sampled, shifted versions of the high resolution image. In many applications, the blurring process, i.e., point spread function (PSF) parameters of the imaging system, is not known. In our blind superresolution algorithm, we rst estimate the PSF parameters from the raw data using the generalized cross-validation method (GCV). To reduce the computational complexity of GCV, we propose eÆcient approximation techniques based on the Lanczos algorithm and Gauss quadrature theory. Blind superresolution experiments are presented to demonstrate the e ectiveness and robustness of our method.