Scribd Logo
Upload

Welcome to Scribd!

  • Abs CCR Clustering and Collaborative Representation

    Uploaded by

    g_31682896
    19 views2 pages
    This document proposes a single image super-resolution algorithm called Clustering and Collaborative Representation (CCR) that clusters low-resolution image features, groups corresponding high-resolution features, and uses multiple projection matrices computed via collaborative representation to map low- to high-resolution feature subspaces. Existing super-resolution methods rely on models and parameters estimated from training data, while the proposed CCR approach clusters training data into subspaces to collect neighbor features and learn local geometry for fast single image super-resolution.

    Original Description:

    ccr

    Original Title

    5. Abs CCR Clustering and Collaborative Representation - Copy

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    This document proposes a single image super-resolution algorithm called Clustering and Collaborative Representation (CCR) that clusters low-resolution image features, groups corresponding high-resolution features, and uses multiple projection matrices computed via collaborative representation to map low- to high-resolution feature subspaces. Existing super-resolution methods rely on models and parameters estimated from training data, while the proposed CCR approach clusters training data into subspaces to collect neighbor features and learn local geometry for fast single image super-resolution.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as docx, pdf, or txt
    19 views2 pages

    Abs CCR Clustering and Collaborative Representation

    Uploaded by

    g_31682896
    This document proposes a single image super-resolution algorithm called Clustering and Collaborative Representation (CCR) that clusters low-resolution image features, groups corresponding high-resolution features, and uses multiple projection matrices computed via collaborative representation to map low- to high-resolution feature subspaces. Existing super-resolution methods rely on models and parameters estimated from training data, while the proposed CCR approach clusters training data into subspaces to collect neighbor features and learn local geometry for fast single image super-resolution.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as docx, pdf, or txt
    of 2

    CCR: Clustering and Collaborative Representation

    for Fast Single Image Super-Resolution

    ABSTRACT

    Clustering and collaborative representation (CCR) have recently been used in fast single
    image super-resolution (SR). In this paper, we propose an effective and fast single image
    super-resolution (SR) algorithm by combining clustering and collaborative representation. In
    particular, we first cluster the feature space of low-resolution (LR) images into multiple LR
    feature subspaces and group the corresponding high-resolution (HR) feature subspaces. The
    local geometry property learned from the clustering process is used to collect numerous
    neighbour LR and HR feature subsets from the whole feature spaces for each cluster centre.
    Multiple projection matrices are then computed via collaborative representation to map LR
    feature subspaces to HR.

    INTRODUCTION

    Single image super-resolution (SR), an important and classical problem in image processing
    and computer vision, aims at restoring a high-resolution (HR) image from its degraded low-
    resolution (LR) measurement, while minimizing visual artefacts as far as possible. SR is a
    typical ill-posed inverse problem since a multiplicity of solutions exist for any input LR pixel
    To mitigate and deal with such a problem, most of SR methods constrain the solution with
    strong prior information, which can be obtained by learning from a set of training data, using
    various regularization methods, and assuming a certain distribution of image features.

    EXISTING SYSTEM

    Classical parametric image processing methods rely on a specific model of the signal
    of interest and seek to compute the parameters of this model in the presence of noise.
    Examples of this approach are presented in diverse problems ranging from denoising to up
    scaling and interpolation. A generative model based upon the estimated parameters is then
    produced as the best estimate of the underlying signal.
    In addition to kernel regression methods we are advocating, there are several other
    effective regression methods such as B-spline interpolation, orthogonal series, cubic spline
    interpolation, and spline smoother. We briefly review some of these methods in this section.
    PRAPOSED SYSTEM

    In this section, we first present how to extract two pools of LR/HR feature sets from a
    training image set and cluster them into numerous subspaces. We then use the clustered
    feature subspaces to group neighbour features, followed by computing multiple projection
    matrices. Finally, the CCR-based SR algorithm is outlined. In the following discussion, we
    denote IL/H , pL/H , and yL/H as an LR/HR image, patch, and feature, respectively. c and F
    denote a cluster center in the LR feature space and its corresponding projection matrix. Bold
    lowercase letters denote vectors. Bold uppercase letters denote regular matrices. Plain
    lowercase letters are used as scalars.
    According to, simple functions mapping from LR feature space to HR feature space can be
    learned for fast SR when abundant and effective exemplars are available in the training
    phase.

    APPLICATION

    Trust-based clustering

    Used in synthetic

    Real time data

    You might also like