Multi-Frame Super-Resolution Reconstruction Based on Anisotropic Markov Random Field Modeling
Zhihui Wei · Dianzi xuebao · 2009
A variational super-resolution reconstruction method is proposed.First of all,a kind of structure-adaptive anisotropic filter is designed based on the recently reported bilateral filtering.It is not only edge-preserving but also cornerpreserving.Then,an anisotropic Markov random field(MRF)model is deduced,which is the improvement of both the classical MRF and bilateral total variation image models.Driven by the anisotropic MRF model,an edge-enhancing super-resolution algorithm is subsequently proposed,simultaneously estimating the high resolution image and the sub-pixel motion among low-resolution frames.The half-quadratic regularization approach and steepest descent are exploited to solve the corresponding minimization functional.Experiment results demonstrate the effectiveness of the proposed approach,both in the visual effect and the peak signal to noise ratio(PSNR)value.