TY - JOUR

T1 - Single-image super-resolution using sparse regression and natural image prior

AU - Kim, Kwang In

AU - Kwon, Younghee

PY - 2010

Y1 - 2010

N2 - This paper proposes a framework for single-image super-resolution. The underlying idea is to learn a map from input low-resolution images to target high-resolution images based on example pairs of input and output images. Kernel ridge regression (KRR) is adopted for this purpose. To reduce the time complexity of training and testing for KRR, a sparse solution is found by combining the ideas of kernel matching pursuit and gradient descent. As a regularized solution, KRR leads to a better generalization than simply storing the examples as has been done in existing example-based algorithms and results in much less noisy images. However, this may introduce blurring and ringing artifacts around major edges as sharp changes are penalized severely. A prior model of a generic image class which takes into account the discontinuity property of images is adopted to resolve this problem. Comparison with existing algorithms shows the effectiveness of the proposed method.

AB - This paper proposes a framework for single-image super-resolution. The underlying idea is to learn a map from input low-resolution images to target high-resolution images based on example pairs of input and output images. Kernel ridge regression (KRR) is adopted for this purpose. To reduce the time complexity of training and testing for KRR, a sparse solution is found by combining the ideas of kernel matching pursuit and gradient descent. As a regularized solution, KRR leads to a better generalization than simply storing the examples as has been done in existing example-based algorithms and results in much less noisy images. However, this may introduce blurring and ringing artifacts around major edges as sharp changes are penalized severely. A prior model of a generic image class which takes into account the discontinuity property of images is adopted to resolve this problem. Comparison with existing algorithms shows the effectiveness of the proposed method.

KW - gradient methods

KW - image matching

KW - image resolution

KW - regression analysis

KW - Kernel ridge regression

KW - gradient descent

KW - kernel matching pursuit

KW - natural image prior

KW - sparse regression

KW - Displays

KW - Energy resolution

KW - Image enhancement

KW - Image resolution

KW - Kernel

KW - Machine learning

KW - Machine learning algorithms

KW - Matching pursuit algorithms

KW - Spatial resolution

KW - Testing

KW - Computer vision

KW - display algorithms.

KW - image enhancement

KW - machine learning

U2 - 10.1109/TPAMI.2010.25

DO - 10.1109/TPAMI.2010.25

M3 - Journal article

VL - 32

SP - 1127

EP - 1133

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 6

ER -