Fractals everywhere
Chaos: making a new science
Information Sciences: an International Journal
On Three Types of Covering-Based Rough Sets
IEEE Transactions on Knowledge and Data Engineering
Comparison of different strategies of utilizing fuzzy clustering in structure identification
Information Sciences: an International Journal
Fractal dimension applied to plant identification
Information Sciences: an International Journal
Research on rough set theory and applications in China
Transactions on rough sets VIII
Fractal dimension and lacunarity of psoriatic lesions: a colour approach
BEBI'09 Proceedings of the 2nd WSEAS international conference on Biomedical electronics and biomedical informatics
Towards video quality metrics based on colour fractal geometry
Journal on Image and Video Processing - Special issue on emerging methods for color image and video quality enhancement
Information Sciences: an International Journal
Hi-index | 0.08 |
Fractal dimensions describe self-similarity (structural complexity) of various phenomena (such as e.g., temporal signals, images). Their determination (through box counting or a correlation method) is inherently associated with the use of information granules--sets. The intent of this study is to generalize the idea to the domain of fuzzy sets and reveal associations between the mechanisms of fractal analysis and granular computing (including fuzzy modeling). First, we introduce the concept itself and discuss the role of fuzzy sets as a vehicle for constructing fractal dimensions. Second, we propose an algorithmic framework necessary to carry out all computing, and experimentally quantify a performance of regression models used to determine fractal dimensions and contrast it with the performance of the fractal models existing in the set-based environment. It is shown that fuzzy set approach produces more consistent models (in terms of their performance). We also postulate a power law of granularity and discuss its direct implications in the form of a variable granularity in fuzzy modeling. In particular, we show how the power law of granularity helps construct mappings between system's variables in rule-based models. Experimental studies involving several frequently used categories of fuzzy sets illustrate the main features of the approach.