The Technique Of Image Coding Based On Fractal Theoretics

In recent years, fractal image compression coding technique, as a new scheme of image compression, has received a great deal of attention and study from many researchers for its desirable properties such as fast decoding, resolution independence of decoded image and high compression ratio. However, there's a especially unsatisfying problem in this method: too long encoding time, mainly because of the considerable number of domain blocks to compare with for each range block in encoding phase, which, in fact, prevent fractal image compression from becoming a practical method for image compression, hence fast encoding has become a hot issue in fractal image compression.In this dissertation, first of all, the theory of fractal image compression coding is introduced; Secondly, the traditional schemes of fractal image compression coding are introduced; Finally, how to accelerate the speed of fractal image compression coding is studied. In order to solve this problem, tow resolutions are proposed in the dissertation:(1) A fast fractal image compression using the improved K-mean clustering is proposed. In this paper, the selection of initial clustering center for K-mean clustering is analyzed, a new initial clustering center selection based on average value and variance is given, and a fast fractal image coding method is proposed. Experimental results show that the proposed coding is a fast and efficient image compression scheme, it can considerably shorten the encoding time, while achieving the same or better decoded image quality.(2)A fast fractal image compression based on fuzzy C-mean clustering is proposed, which can search out the best-matched block to an input range block with a reduced search. Firstly, the origin image is split into range blocks and domain blocks, and 8 elementary transform are performed on domain blocks to obtain the domain blocks group. Secondly, all range blocks and domain blocks group are clustered by fuzzy C-means clustering (FCM). Finally, domain block transform with the biggest fuzzy membership are encoded. Experimental results show that the proposed image coding is a fast and efficient image compression scheme; it can considerably shorten the encoding time, while achieving the same or better decoded image quality.