Fractal Analysis for Reduced Reference Image Quality Assessment
In this paper, multi fractal analysis is adapted to reduced-reference image quality assessment (RR-IQA). A novel RR-QA approach is proposed, which measures the difference of spatial arrangement between the reference image and the distorted image in terms of spatial regularity measured by fractal dimension. An image is first expressed in Log-Gabor domain. Then, fractal dimensions are computed on each Log-Gabor sub band and concatenated as a feature vector. Finally, the extracted features are pooled as the quality score of the distorted image using _1 distance. Compared with existing approaches, the proposed method measures image quality from the perspective of the spatial distribution of image patterns. The proposed method was evaluated on seven public benchmark data sets. Experimental results have demonstrated the excellent performance of the proposed method in comparison with state-of-the-art approaches.
To quantify how much an image is affected by degradation, a metric is required ti evaluate how good an image is in visual meaning for human visual system (HVS).This leads to the research on image quality assessment (IQA).
As the ultimate solution, subjective IQA has its advantages on the reliability and consistency with human, for the quality is directly quantified by observes. Nevertheless, the practicality of subjective IQA is very limited because it is expensive and time-consuming.
DISADVANTAGES OF EXISTING SYSTEM:
Histogram-based method cannot correctly reflect the degree of the distortion
In this paper implemented by using Log-Gabor representation and fractal analysis. The former aims to create a complete basis of visual perception, while the latter can encode spatial information in the form of the geometrical distribution of visual data.
Specifically, the proposed RR-IQA feature, called spectrum of spatial regularity (SSR), characterizes the spatial distribution of image structures based on fractal analysis. The spatial-frequency components of image are first extracted by Log-Gabor filtering. Then fractal dimension is used to measure the spatial regularity of the arrangements in each Log-Gabor sub band. Finally all the computed fractal dimensions are collected as a feature vector. By using fractal analysis that has a strong correlation with HVS, the image structures are well encoded and the difference of their spatial arrangements between images can be well characterized.
Unlike previous work based on fractal analysis, in our method fractal analysis is working on the visual perceptive space instead of image space due to the fact that HVS is a limited-bandwidth system which is sensitive to specific spatial frequencies. Our approach was evaluated on seven public IQA benchmark databases using five evaluation criteria. The competitive results achieved demonstrate that our method performs on par with the state-of-the-art approaches.
ADVANTAGES OF PROPOSED SYSTEM:
A useful RR-IQA metric is expected to achieve higher prediction accuracy while using less information of the reference image.
Competitive performance among state-of-the-arts, consistent performance across different types of distortion, high ratio of accuracy over data rate, moderate and acceptable computational cost.
System : Pentium IV 2.4 GHz.
Hard Disk : 40 GB.
Floppy Drive : 44 Mb.
Monitor : 15 VGA Colour.
Ram : 512 Mb.
Operating system : Windows XP/7.
Coding Language : MATLAB
Tool : MATLAB R2013A
Yong Xu, Delei Liu, Yuhui Quan, and Patrick Le Callet, “Fractal Analysis for Reduced Reference Image Quality Assessment”, IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 24, NO. 7, JULY 2015.