Abstract
One method of solving the single-image super-resolution problem is to use Heaviside functions. This has been done previously by making a binary classification of image components as “smooth” and “non-smooth”, describing these with approximated Heaviside functions (AHFs), and iteration including l1 regularization. We now introduce a new method in which the binary classification of image components is extended to different degrees of smoothness and non-smoothness, these components being represented by various classes of AHFs. Taking into account the sparsity of the non-smooth components, their coefficients are l1 regularized. In addition, to pick up more image details, the new method uses an iterative refinement for the residuals between the original low-resolution input and the downsampled resulting image. Experimental results showed that the new method is superior to the original AHF method and to four other published methods.
Highlights
Image super-resolution (SR) is to generate or recover high-resolution (HR) images from one or multiple low-resolution (LR) images
Single image SR based on approximated Heaviside functions and iterative refinement represented by multiple classes of AHFs with sharp edges
The single image super-resolution via iterative AHF method (AHFM) proposed in [7] for image super-resolution gives a selection of sharp-related terms, which are measured from the LR image input and apply them to fine grids to generate the HR image
Summary
Image super-resolution (SR) is to generate or recover high-resolution (HR) images from one or multiple low-resolution (LR) images. Yang et al [12] apply the theory of sparse coding to SR problems effectively This method jointly trains two dictionaries for LR and HR image patches, and they could use the LR dictionary to generate sparse representations of the LR input to obtain the corresponding HR output. We study an effective single-frame SR approach, which is an improvement of the so called approximated Heaviside functions method (AHFM) proposed in [7] It shows that the underlying image, can be viewed as a instensity function, which can be approximately represented by two classes of AHFs. Deng et al cast the image super-resolution problem as an intensity function estimation problem. Single image SR based on approximated Heaviside functions and iterative refinement represented by multiple classes of AHFs with sharp edges. A brief review of the method based on approximated Heaviside functions can be found from [7]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
