In this study, two methods have been suggested for breaking the long encoding time problem of the encoding process in Fractal Image Compression (FIC). The First, called Zero Mean Intensity Level (ZMIL), which is based on using an unconventional affine parameter (namely the range block mean) that has better properties than the conventional offset parameter. As result, it is found that ZMIL gives a high value of compression ratio (around 18.9% additional compression value), with a high reconstructed image quality and a reduction in the encoding time of about 27% in comparison with the traditional FIC. The second, called the speed up (ZMIL) method which is responsible for the reduction of the number of domain blocks needed to be IFS matched with tested range block by eliminating the symmetry orientations and Reduction Domain Image Size (RDIS). From the results of this method, the encoding time will be reduced to only one second (i.e., reduction about 95% of the time required in the full cases), with a higher increase in the value of compression ratio and still has a good reconstructed image quality (PSNR).
[1]
Lyman P. Hurd,et al.
Fractal image compression
,
1993
.
[2]
Hong Yan,et al.
Adaptive Fractal Image Compression
,
1994
.
[3]
Y. Fisher.
Fractal image compression: theory and application
,
1995
.
[4]
Hannes Hartenstein,et al.
Adaptive partitionings for fractal image compression
,
1997,
Proceedings of International Conference on Image Processing.
[5]
L. Soberano.
THE MATHEMATICAL FOUNDATION OF IMAGE COMPRESSION By
,
2000
.
[6]
李 信行.
Parallel Processing Architecture for Fractal Image Compression
,
2000
.
[7]
Dietmar Saupe,et al.
Combining fractal image compression and vector quantization
,
2000,
IEEE Trans. Image Process..
[8]
Ming Hong Pi,et al.
Fast fractal image encoding based on adaptive search
,
2001,
IEEE Trans. Image Process..
[9]
R. S. D. Wahida Banu,et al.
Adaptive fractal image compression using PSO
,
2010,
Biometrics Technology.