English abstract
With the advances in the information era, there is a rapid rise in the volume of digital data generated by different types of applications ranging from social
media networks all the way to satellite and medical images. Data compression
plays an important role to solve the storage space requirement and data
transmission time constraints. Fractal coding is a lossy image compression
technique, which encodes the image in a way that would require less storage space by using the self-similar nature of the image. The main drawback of fractal compression is the high encoding time. This is due to the hard tasks of finding all fractals during the partition step and the search for the best match of fractals. Researchers have proposed different methods to overcome this drawback including classification schemes, which are mostly used to speed up the encoding process. Lately, GPUs (Graphical Processing Unit) have been exploited to implement fractal image compression algorithms due to their high computational power.
The prime aim of this research is to experimentally evaluate the performance of Fisher classification scheme that is widely used to reduce the encoding
time of fractal images by limiting the search for the best match of fractals. CUDA (Compute Unified Device Architecture) has been used to design and implement the Fishers classification scheme to exploit the computational power available in the GPUs.
Encoding time, compression ratio and peak signal-to-noise ratio was used as metrics to compare the performance of the developed algorithm. Eigh images with different sizes have been used (512x512, 1024x1024 and 2048x2048)
for the experiments. The conducted experiments showed that a speedup of 6.4x was achieved in some images using NVIDIA GeForce GT 660M GPU.
