A novel regularised image super-resolution algorithm is proposed, building on the emerging cosparse or analysis sparse prior models, which are important complementary alternatives to the widely used synthesis sparse counterpart. Moreover, to achieve adaptivity to the varying local structures of natural images, the patch space is partitioned into meaningful subspaces by clustering and learn analysis sub-dictionary for each cluster are partitioned, which are performed online and iteratively based solely on the current available image information, for maximum generality and flexibility. In addition, non-local feature self-similarity is incorporated for further reconstruction quality enhancement. Experimental results show that the proposed approach gives favourable results with respect to the state-of-the-art methods.

Volume scattering diffusers (VSDs) with different thicknesses were fabricated for speckle reduction investigation in a two-diffuser system. The VSDs were obtained by spin-coating the mixture of SiO2 microspheres and SU-8 photoresist, where the monodispersed SiO2 microspheres were synthesized by the Stöber method with an average diameter of 1.43 µm. The Mie scattering effect of the VSDs was observed owing to the refractive index difference between the SiO2 microspheres and the SU-8 photoresist. The speckle reduction effect of the two-diffuser system comprising cascaded stationary and moving VSDs was experimentally studied. The result can be used in designing more effective speckle reduction techniques in laser displays.

No abstract