Signal Processing - March 2016 - 12

Image Super-Resolution
Via Sparse Representation

https://www.graphicstock.com/stock-image/digital-female-140220

Yang, J.; Wright, J.; Huang, T.S.; Ma, Y.
This paper presents an approach to single-image super-resolution based upon
sparse signal representation of low- and
high-resolution patches. Coefficients
are determined for each patch of the
low-resolution input, and then used to
generate the high-resolution output.
November 2010

Vector Sparse Representation of
Color Image Using Quaternion
Matrix Analysis
Xu, Y.; Yu, L.; Xu, H.; Zhang H.;
Nguyen, T.
The proposed model represents the color
image as a quaternion matrix, where a
quaternion-based dictionary learning
algorithm is presented using the K-quaternion singular value decomposition
(QSVD) (generalized K-means clustering for QSVD) method. It conducts the
sparse basis selection in quaternion
space, which uniformly transforms the
channel images to an orthogonal color
space. In this new color space, the inherent color structures can be completely
preserved during vector reconstruction.
April 2015

Image Quality Assessment for
Fake Biometric Detection: Application
to Iris, Fingerprint, and Face
Recognition
Galbally, J.; Marcel, S.; Fierrez J.
The proposed system attempts to enhance
the security of biometric recognition
frameworks by adding a liveness assessment in a fast, user-friendly, and nonintrusive manner. Twenty-five general
image-quality features are extracted from
the authentication image to distinguish
between legitimate and imposter samples
for fingerprint, iris, and two-dimensional
face biometrics.
Februrary 2014

Vector-Valued Image Processing
by Parallel Level Sets
Ehrhardt, M.J.; Arridge, S.R.
This paper considers the components
of an image as a vector based on the
angle between the spatial gradients
of their channels. By minimizing
large angles parallel level sets are
obtained and used for demosaicking.
January 2014

Face Recognition Across
Non-Uniform Motion Blur,
Illumination, and Pose
Punnappurath, A.; Rajagopalan, A.N.;
Taheri, S.; Chellappa, R.; Seetharaman, G.
A methodology for face recognition
in the presence of space-varying
motion blur comprising of arbitrarily
shaped kernels is presented. The
authors model the blurred face as a
convex combination of geometrically
transformed instances of the focused
gallery face and show that the set of
all images obtained by nonuniformly
blurring a given image forms a convex set. The framework is then
extended to handle illumination varia-

IEEE Signal Processing Magazine

March 2016

tions by exploiting the fact that the set
of all images obtained from a face
image by nonuniform blurring and
changing the illumination forms a
biconvex set.
July 2015

Active Contours
Without Edges
Chan, T.F.; Vese, L.A.
This paper presents a model to detect
o b j e c t s u s i n g c u r ve evo l u t i o n .
Numerical comparisons show the
advantage of a stopping criterea
based on the segmentation rather
than the gradient.
February 2001

Weighted Guided
Image Filtering
Li, Z.; Zheng, J.; Zhu, Z.; Yao, W.; Wu, S.
The weighted guided image filter
incorporates an edge-aware weighting into existing guided image filter
to address the problem of halo artifacts. The filter is applied to detail
enhancement, haze removal, and
image fusion.
January 2015

Image Denoising Via Sparse
and Redundant Representations
over Learned Dictionaries
Elad, M.; Aharon, M.
This paper uses either a corrupted
image itself or a high-quality image
data base to learn dictionaries of small
patches generated via the K-SVD
algorithm. A global image prior is
used to extend small patches to the
entire image.
December 2006

https://www.graphicstock.com/stock-image/digital-female-140220

Table of Contents for the Digital Edition of Signal Processing - March 2016