Self-adaptive blind super-resolution image reconstruction

Super-resolution (SR) image reconstruction is a rapidly developing area in image processing. Especially, blind SR can generate high space resolution image without requiring priori information of the point spread function (PSF). In this paper, we propose a self-adaptive blind super-resolution image reconstruction approach which is based on multiple images. Our method can adaptively choose the parameter of regularization term. The quality of resulting image especially in the respect of edge-preserving property is better than approaches such as maximum a posteriori estimation (MAP) tested with practical examples. We employ Lorentzian function as spread coefficient and partial differential function as regularization term of resulting image. A generalized version of the eigenvector-based alternating minimization (EVAM) constraint is used to regularize PSF and estimate resulting image and PSF simultaneously. In addition, in order to achieve self-adaptive regularization terms parameter choosing, we also present a new robust no-reference image quality assessment method which provides blurring and ringing effect assessment value as feedback.