Volume 8 Issue 1 - January 2017

  • 1. Super resolution reconstruction of image gradient profile sharpness

    Authors : Minal Chandurkar, Roshni Khedgaonkar

    Pages : 329-335

    DOI : http://dx.doi.org/10.21172/1.81.042

    Keywords : Super Resolution Single ImageGradient Profile SharpnessProfile dictionary

    Abstract :

    Single Image super resolution is an active and classic image processing problem, which aims to reconstruct a high resolution (HR) single image from a low resolution input image. Due to the several use of profile dictionary under-determined nature of this type of problem, an effective image prior is necessary to make the problem solvable, and to improve the quality of reconstruct super resoluted image. In this paper image super resolution algorithm is proposed based on gradient profile sharpness. Gradient Profile Sharpness (GPS) is an edge sharpness matrix which is extracted from two gradient description models, i.e. a Gaussian mixture model for the description of different kind of gradient profile. The proposed approach will generate superior HR image with better visual quality, lower reconstruction error, and acceptable computation time less than the existing algorithm. To improve the HR image pixel quality, we will be use some filters. And compare the PSNR values to the existing GMM method.

    Citing this Journal Article :

    Minal Chandurkar, Roshni Khedgaonkar, "Super resolution reconstruction of image gradient profile sharpness", https://www.ijltet.org/journal_details.php?id=910&j_id=3597, Volume 8 Issue 1 - January 2017, 329-335, #ijltetorg