With recent advances in technology and the spread of computer and Internet use, there has been a significant increase in the use of digital images. The large number of bits needed to represent pictures, and the resulting image size has created problems in the processing, transmission and storage of digital media. This has led to the need for new and more efficient compression processes to reduce the amount of data necessary to represent the image while maintaining good quality compared to the original image. Since only limited research has been conducted in regard to how these new formats have taken into account the human visual system (HVS), it is necessary to define these formats, compare them, and categorize them based on the user perceived quality of the compressed images. In this paper, three image standards, JPEG 2000, HD Photo and JPEG, are compared in relationship to image quality and bit-rate. The study utilizes a simple evaluation methodology to compare the compression performance between the three image formats. Both subjective mean opinion score (MOS) and objective peak signal-to-noise ratio (PSNR) and the perceptual structural similarity (SSIM) were identified as distortion metrics and were used for comparisons and subsequent analysis. The study concluded that the effects of image compression differ among various types of images and that human evaluation of image quality is not consistent with automated objective measurements, but has shown higher correlation using the SSIM HVS-based method. In general, the analysis shows that overall performances in terms of image quality are quite comparable for the three coding formats, with JPEG 2000 outperforming JPG and HD Photo in bit-rate, especially at lower compression levels. The research also shows that the SSIM is a better indicator of perceived image quality than the popular PSNR metric, especially for highly compressed images.
[1]
D. Amnon Silverstein,et al.
The relationship between image fidelity and image quality
,
1996,
Proceedings of 3rd IEEE International Conference on Image Processing.
[2]
Zhou Wang,et al.
Modern Image Quality Assessment
,
2006,
Modern Image Quality Assessment.
[3]
Alan C. Bovik,et al.
Mean squared error: Love it or leave it? A new look at Signal Fidelity Measures
,
2009,
IEEE Signal Processing Magazine.
[4]
Wilfried Osberger,et al.
Perceptual vision models for picture quality assessment and compression applications
,
1999
.
[5]
Zhou Wang.
Objective Image/Video Quality Measurement A Literature Survey
,
1998
.
[6]
Mohamed Rahoma El-Hesnawi.
Subjective analysis of image coding errors
,
2002
.
[7]
Francesca De Simone,et al.
A comparative study of color image compression standards using perceptually driven quality metrics
,
2008,
Optical Engineering + Applications.
[8]
Eero P. Simoncelli,et al.
Image quality assessment: from error visibility to structural similarity
,
2004,
IEEE Transactions on Image Processing.
[9]
Patrick Le Callet,et al.
An image quality assessment method based on perception of structural information
,
2003,
Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).
[10]
Paul S. Fisher,et al.
Image quality measures and their performance
,
1995,
IEEE Trans. Commun..