The output quality of an image filter for reducing noise without damaging the underlying signal, strongly depends on the
accuracy of the noise model in characterizing the noise introduced by the acquisition device. In this paper we provide
a solution for characterizing signal dependent noise injected at shot time by the image sensor. Different fitting models
describing the behavior of noise samples are analyzed, with the aim of finding a model that offers the most accurate
coverage of the sensor noise under any of its operating conditions. The noise fitting equation minimizing the residual error
is then identified. Moreover, a novel algorithm able to obtain the noise profile of a generic image sensor without the need of
a controlled environment is proposed. Starting from a set of heterogeneous CFA images, by using a voting based estimator,
the parameters of the noise model are estimated.
Accurate noise level estimation is essential to assure good performance of noise reduction filters. Noise contaminating
raw images is typically modeled as additive white and Gaussian distributed (AWGN); however raw images
are affected by a mixture of noise sources that overlap according to a signal dependent noise model. Hence, the
assumption of constant noise level through all the dynamic range represents a simplification that does not allow
precise sensor noise characterization and filtering; consequently, local noise standard deviation depends on signal
levels measured at each location of the CFA (Color Filter Array) image.
This work proposes a method for determining the noise curves that map each CFA signal intensity to its
corresponding noise level, without the need of a controlled test environment and specific test patterns. The
process consists in analyzing sets of heterogeneous raw CFA images, allowing noise characterization of any image
sensor. In addition we show how the estimated noise level curves can be exploited to filter a CFA image, using
an adaptive signal dependent Gaussian filter.
An objective image quality metric can be used to compare the output of different image processing algorithms, but objective measures are not always well correlated with subjective image quality assessment; the latter implies the use of human observers, thus objective methods able to emulate the Human Visual System (HVS) better than the classical measures are preferred. In this paper a full reference objective metric, based on perceptual criteria and oriented to demosaiced images is proposed.
The basic idea is to model the main artifacts produced by the interpolation process, taking into account the HVS sensibility to the typical aliasing and the zipper defects. The proposed technique has been compared to the DE94 CIELAB metric. Furthermore, two subjective tests have been performed; one relative to the color aliasing artifact and one to the zipper effect. The experimental results highlight that the quality scores obtained by the proposed measures have a similar trend to the DE94 CIELAB metric. Moreover, subjective tests are in accordance with the obtained results.
This technique is useful to evaluate the quality of the interpolation techniques implemented in the image processing pipeline of different digital still cameras.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.