Abstract
This paper presents an adaptive image interpolation method using a pixel-based neighbor embedding which is modified from the patch-based neighbor embedding of contemporary super resolution algorithms. Conventional interpolation methods for high resolution detect at least 16-directional edges in order to remove zig-zaging effects and selectively choose the interpolation strategy according to the direction and value of edge. Thus, they require much computation and high complexity. In order to develop a simple interpolation method preserving edge’s directional shape, the proposed algorithm adopts the simplest Haar wavelet and suggests a new pixel-based embedding scheme. First, the low-quality image but high resolution, magnified into 1 octave above, is acquired using an adaptive 8-directional interpolation based on the high frequency coefficients of the wavelet transform. Thereafter, the pixel embedding process updates a high resolution pixel of the magnified image with the weighted sum of the best matched pixel value, which is searched at its low resolution image. As the results, the proposed scheme is simple and removes zig-zaging effects without any additional process.
Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
