image proceesing

1. BLIND IMAGE QUALITY
ASSESSEMENT: A
NATURAL SCENE
STATISTICS APPROACH
IN THE DCT DOMAIN

2. CONTENTS:-
 Abstract
 Introduction
 Blind image
 Natural scene statistics
 Discrete cosine transforms
 Generalized Gaussian density
 Blind image quality assessment
 No reference image quality


3. ABSTRACT:-
 We propose an efficient, general-purpose,
distortion-agnostic, blind/no-referenceimage
quality assessment (NR-IQA) algorithm based
on a natural scene statistics model of discrete
cosine transform (DCT) coefficients.
 The algorithm is computationally appealing,
given the availability of platforms optimized
for DCT computation. We propose a
generalized parametric model of the extracted
DCT coefficients. The parameters of the
model are utilized to predict image quality
scores.

4. Contd…
 The resulting algorithm, which we name
BLIINDS-II, requires minimal training and
adopts a simple probabilistic model for score
prediction.
 When tested on the LIVE IQA database,
BLIINDS-II is shown to correlate highly with
human visual perception of quality, at a level
that is even competitive with the powerful full-
reference SSIM index.

5. Introduction:-
 Measurement of image or video quality is crucial for many
image-processing algorithms, such as acquisition,
compression, restoration,enhancement, and reproduction.
Traditionally, image quality assessment (QA) algorithms
interpret image quality as similarity with a "reference" or
"perfect" image.
 The obvious limitation of this approach is that the reference
image or video may not be available to the QA algorithm.
The field of blind, or no-reference, QA, in which image
quality is predicted without the reference image or video, has
been largely unexplored, with algorithms focussing mostly on
measuring the blocking artifacts.
 Emerging image and video compression technologies can
avoid the dreaded blocking artifact by using various
mechanisms, but they introduce other types of distortions,
specifically blurring and ringing.

6. Contd…
 In this paper, we propose to use natural scene
statistics (NSS) to blindly measure the quality of
images compressed by JPEG2000 (or any other
wavelet based) image coder.
 We claim that natural scenes contain nonlinear
dependencies that are disturbed by the compression
process, and that this disturbance can be quantified
and related to human perceptions of quality.
 We train and test our algorithm with data from human
subjects, and show that reasonably comprehensive
NSS models can help us in making blind, but
accurate, predictions of quality. Our algorithm
performs close to the limit imposed on useful
prediction by the variability between human subjects.

7.  Blind image: Image deposed, embossed
or stamped, but not printed with ink or foil.
 Natural scene characteristics: Scene
statistics is a discipline within the field of
perception. It is concerned with the
statistical regularities related to scenes. It
is based on the premise that a perceptual
system is designed to interpret scenes.

8. The Discrete Cosine
Transform (DCT):-
 The discrete cosine transform (DCT) helps separate the image into
parts (or spectral sub-bands) of differing importance (with respect to
the image's visual quality). The DCT is similar to the discrete
Fourier transform: it transforms a signal or image from the spatial
domain to the frequency domain.

9. Over view of the method:-
 We will refer to undistorted images captured by imaging devices
that sense radiation from the visible spectrum as natural scenes,
and statistical models built for undistorted natural scenes as NSS
models. Deviations from NSS models, caused by the introduction
of distortions to images, can be used to predict the perceptual
quality of the image. The model-based NSS-IQA approach
developed here is a process of feature extraction from the image,
followed by statistical modeling of the extracted features.
 Purely NSS-based IQA approaches require the development of a
distance measure between a given distorted test image and the
NSS model.
 This leads to the question of what constitutes appropri- ate and
perceptually meaningful distance measures between distorted
image features and NSS models. The Kullback–Leibler
divergence [21] as well as other distance measures have been
used for this purpose, but no perceptual justification has been
provided for its use.

10. Contd…
 Our approach relies on the IQA algorithm
learning how the NSS model parameters vary
across different perceptual levels of image
distortion.
 The algorithm is trained using features derived
directly from a generalized parametric
statistical model of natural image DCT
coefficients against various perceptual levels
of image distortion. The learning model is then
used to predict perceptual image quality
scores.

11. No-reference image quality
assessment based on DCT domain
statistics:-
 This paper proposes a no-reference quality assessment
metric for images subject to quantization noise in block-
based DCT (discrete cosine transform) domain, as those
resulting from JPEG or MPEG encoding.
 The proposed method is based on natural scene
statistics of the DCT coefficients, whose distribution is
usually modeled by a Laplace probability density
function, with parameter λ.
 A new method for λ estimation from quantized coefficient
data is presented; it combines maximum-likelihood with
linear prediction estimates, exploring the correlation
between λ values at adjacent DCT frequencies.

12. Contd…
 The resulting coefficient distributions are then used
for estimating the local error due to lossy encoding.
Local error estimates are also perceptually weighted,
using a well-known perceptual model by Watson.
 When confronted with subjective quality evaluation
data, results show that the quality scores that result
from the proposed algorithm are well correlated with
the human perception of quality.
 Since no knowledge about the original (reference)
images is required, the proposed method resembles a
no-reference quality metric for image evaluation.

13. Generalized Gaussian Model
Shape Parameter:-
 We deploy a generalized Gaussian model of the non-DCDCT
coefficients from nxn blocks. The DC coefficient does not convey
structural information about the block, including it neither increases
nor decreases performance.
 The generalized Gaussian density in (1) is parameterized by mean μ,
scale parameter β, and shape parameter γ . The shape parameter γ is
a model-based feature that is computed over all blocks in the image.
 The shape parameter quality feature is pooled in two ways. First, by
computing the lowest 10th percentile average of the local block shape
scores (γ ) across the image

14. Contd…
 Thiskind of “percentile pooling” has been observed to
result in improved correlations with subjective perception
of quality [23], [45]. Percentile pooling is motivated by the
observation that the “worst” distortions in an image
heavily influence subjective impressions. We choose
10% as a round number to avoid the possibility of
training. In addition, we compute DCT coefficients, three
bands.
 The 100th percentile average (ordinary sample mean) of
the local γ scores across the image. Using both 10% and
100% pooling helps inform the predictor whether the
distortions are uniformly annoying over space or exhibit
isolated perceptually severe distortions.

15. Blind Image Quality
Assessment
 Human visual system usually does not
need any reference to determine the
subjective quality of a target image
 Distinction between fidelity and quality.
 Edge Sharpness Level
 Random Noise Level
 Structured Noise Level

16. Edge Sharpness Level

17. Random Noise Level
 Impulse Noise
 Additive White Gaussian Noise

18. Structured Noise Level
 Block Artifact
 Ringing Artifact