Graphics Reference
In-Depth Information
Image Quality Assessment Based on SIFT and SSIM
Wenjun Lu, Congli Li, Yongchang Shi, and Xiaoning Sun
New Star Research Institute of Applied Technology, Hefei 230031, China
{Wenjun.lu2013,shi756603491,sunxiaoning0117}@gmail.com,
lcliqa@163.com
Abstract.
Image quality assessment (IQA) aims to provide computational mod-
els to measure the image quality consistently with subjective assessments. The
SSIM index brings IQA from pixel-based to structure-based stage. In this paper,
a new similarity index based on SIFT features (SIFT-SSIM) for full reference
IQA is presented. In the algorithm, proportion of matched features in extracted
features of reference image and structural similarity are combined into a com-
prehensive quality index. Experiments on LIVE database demonstrate that
SIFT-SSIM is competitive with most of state-of-the-art FR-IQA metrics, and it
can achieve higher consistency with the subjective assessments in some distor-
tion types.
Keywords:
Image Quality Assessment, Full Reference, Structural Similarity,
Space Invariant Feature Transform.
1
Introduction
1.1
Image Quality Assessment
Image quality is an important indicator of varieties of image processing algorithms
and optimizing system parameters. To establish an effective mechanism for image
quality assessment has very important significance in image acquisition, encoding,
network transmission and other areas [1]. In recent years, with the development of
image processing technology, image quality assessment has attracted wide attention
from researchers. There are many domestic and foreign research institutions and
commercial companies join to research [2].
Image quality assessment can be divided into objective and subjective assessment
methods, the former is by virtue of subjective perception experiments to evaluate the
quality of an object; latter bases on model to give quantitative indicators, and to simu-
late perception mechanisms of human visual system to measure image quality.
Relative to subjective quality assessment, objective quality assessment has become a
research focus for advantages of simple, low cost, easy to parse and embed. Combina-
tion subjective with objective assessment method was focused on applications, and it
uses results of subjective assessment to correct model parameters of objective quality
assessment.
