TIFS: a hybrid scheme integrating partitioned iterated function system and linear transforms

Since its introduction, fractal theory has been exploited in many research topics such as image processing, pattern recognition and image coding, because of the intrinsic properties which make fractals highly suited to tasks such as segmentation, feature extraction and indexing, just to name a few. Unfortunately, they are based on a strong asymmetric scheme, consequently suffering from very high coding times. On the other hand, linear transfoms are quite time balanced, allowing them to be usefully exploited in real-time applications, but they do not provide comparable performance with respect to the image quality for high bit-rates. Here, different levels of embedding linear transforms in the fractal coding scheme are investigated. Experimental results have been organised to point out the contribution of each embedding step to the objective quality of the decoded image.