Noisy Single Image Super-Resolution Based on Local Fractal Feature Analysis

Generally, most existing super-resolution (SR) methods do not consider noise, which treats SR reconstruction and denoising as two separate problems performs separately. However, noise is inevitably introduced in the imaging process. Based on analysis of degraded model, this paper, interpolation are modeled to estimate noiseless missing images under same framework. By applying local fractal dimension (LFD) into image feature analysis, a noisy single-image method proposed. For each image, we first construct rational model containing scaling factors, can effectively maintain inherent properties data. Furthermore, original structure be well preserved by model. Considering characteristics factors calculated basis LFDs. Then, through further interpolated based LFD proposed for recovering image. Finally, high-quality high-resolution obtained. Experimental results demonstrate that our outperforms state-of-the-art both quantitatively qualitatively.