Region-based Super-Resolution Reconstruction Using Parallel Programming

This paper presents region-based super-resolution reconstruction approach using a parallel programming for minimizing interpolation artifacts. In the proposed algorithm of regularized iterative SR, two different types of constraints are incorporated, namely (i) edge reconnection and (ii) orientation adaptive filtering. By combining these two constraints the proposed SR algorithm can successfully remove aliasing artifact and, as a result, produce edge-preserving high-resolution (HR) images. Experimental results show that the proposed algorithm is more effective than traditional methods