Double sparsity for multi-frame super resolution

Toshiyuki Kato, Hideitsu Hino, Noboru Murata

    Research output: Contribution to journalArticle

    5 Citations (Scopus)

    Abstract

    A number of image super resolution algorithms based on the sparse coding have successfully implemented multi-frame super resolution in recent years. In order to utilize multiple low-resolution observations, both accurate image registration and sparse coding are required. Previous study on multi-frame super resolution based on sparse coding firstly apply block matching for image registration, followed by sparse coding to enhance the image resolution. In this paper, these two problems are solved by optimizing a single objective function. The proposed formulation not only has a mathematically interesting structure, called the double sparsity, but also yields comparable or improved numerical performance to conventional methods.

    Original languageEnglish
    Pages (from-to)115-126
    Number of pages12
    JournalNeurocomputing
    Volume240
    DOIs
    Publication statusPublished - 2017 May 31

      Fingerprint

    Keywords

    • Dictionary Learning
    • Double Sparsity
    • Image Super Resolution
    • Sparse Coding

    ASJC Scopus subject areas

    • Computer Science Applications
    • Cognitive Neuroscience
    • Artificial Intelligence

    Cite this