Granular Art - principles of knowledge-based image understanding
The last couple of decades has seen the application of digital computers in all aspects
of human activity. The power of computer-based information processing has been harnessed
for modelling of physical and social systems, communication, process control and decision-making
to mention just a few. It is through the application of computers to modelling ever more complex systems
that the limitations of algorithmic (as opposed to human) information processing became apparent.
We are now entering a new and exciting computing paradigm; that of granular information processing,
entailing generation of information abstractions, their empirical verification and their use in building
models that attempt to balance the generality and uncertainty at different levels of abstraction.
In other words, the human-centric perspective on information processing and knowledge
generation became an inspiration and provided impetus for the re-evaluation of computers as
tools, which critically depend on external empirical input if they are to facilitate a transition from
information to knowledge. Having appreciated the key role of empirical verification
we begin to see the synergy between Sciences, Arts and Computing.
It is clear that all these activities, that have been frequently considered as disjoined,
share a characteristic of generating information abstractions and verifying
them through the reference to the real-world. While the abstractions generated in
different domains can be very different,
the methodological underpinning of their verification is shared universally. This is why our initial
investigation focused on the development of formal mathematical framework for processing information abstractions
(information granules), that of Granular Computing before progressing to Granular Art.
Granular Art is a combined artistic and scientific endeavor aimed at capturing
and expressing the complex artistic messages through a computer-aided abstraction
(granulation) that operates at various cognitive levels. The processing of geometric forms,
colours and textures into conceptual granules is a first step in distilling
the essence of an artistic artefact. This continues at the level of social/cultural/historical
context informing particular responses to texture, contrast, choice of colour, juxtaposition
of forms etc.. Finally, the granulation of the semantical content of complementary
natural language messages, from the artist and the recipients of a given art-form,
helps to cristalise the meaning of the artistic message. The research method of Granular
Art is therefore twofold, granulation of simple abstractions into more complex ones (and their
subsequent verification) and a de-construction of the narrative also followed by
the verification of the de-constructed concepts.
The Granular Art Exhibition in the Gallery EXTRAVAGANCE
is a pioneering attempt to explore synergies between Arts and Computational Intelligence.
The granular computing framework, ensures the preservation of the semantic currency
of complex, multi-layered artistic messages, and facilitates crystallisation
of artists' intuition about human processing of
information abstractions. It is interesting to note that the interpretation of visual art
in its social/cultural/historical context generates information granules that
transcend the visual domain and resonate with our deeper knowledge of the world.
- Bargiela, A., Pedrycz, W., Granular computing: an introduction, Springer, 2003
- Bargiela, A., Pedrycz, W., (Eds.) Human-centric information processing through granular modelling, Studies in Computational Intelligence 182, Springer Berlin Heidelberg, 2009, (doi: 10.1007/978-3-540-92916-1)
- Bargiela A., Pedrycz W., Hirota K., Data granulation through optimization of similarity measure, Archives of Control Sciences, 12, 4, 469-491, 2002
- Pedrycz, W., Bargiela, A., Granular clustering: a granular signature of data, IEEE Trans. on Systems Man and Cybernetics, SMC-B, 32, 2, April 2002, 212-224, (doi: 10.1109/3477.990878 )
- Bargiela, A., Pedrycz, W., Recursive information granulation: Aggregation and interpretation issues, IEEE Trans. on Systems Man and Cybernetics SMC-B, 33, 1, 17, 2003, 96-112. (10.1109/TSMCB.2003.808190)
- Bargiela A., Pedrycz W., Hirota K., Granular prototyping in fuzzy clustering, IEEE Transactions on Fuzzy Systems, 12, 5, 2004, 697-709 (doi: 10.1109/TFUZZ.2004.834808)
- Bargiela, A., Pedrycz, W., Granular mappings, IEEE Transactions on Systems Man and Cybernetics SMC-A, vol. 35, 2, March 2005, 288-301 (doi: 10.1109/TSMCA.2005.843381)
- Bargiela, A., Pedrycz, W., A model of granular data: a design problem with the Tchebyschev FCM, Soft Computing, 9, 3, March 2005, 155-163 (doi: 10.1007/s00500-003-0339-2)
- Bargiela A., Homenda W., Information structuring in natural language communication Ð syntactical approach, Journal of Intelligent and Fuzzy Systems, 17(6), 2006, 575-582.
- Bargiela, A., Pedrycz, W., The roots of granular computing, Proceedings of 2006 IEEE International Conference on Granular Computing, 806-809
- Bargiela, A., Pedrycz, W., Toward a theory of Granular Computing for human-centred information processing, IEEE Trans. on Fuzzy Systems, vol. 16, 2, 2008, 320-330. (doi: 10.1109/TFUZZ.2007.905912)
- Pedrycz W., Bargiela A., Fuzzy clustering with semantically distinct families of variables: descriptive and predictive aspects, Pattern Recognition Letters, 31(13), 1952-1958, 2010. (doi: 10.1016/j.patrec.2010.06.016)
- Pedrycz, W., Bargiela, A., An Optimisation of allocation of information gramularity in the interpretation of data structures, IEEE Transactions on Systems Man and Cybernetics Part B , 42(3), 2012, 582-590 (doi: 10.1109/TSMCB.2011.2170067)
- Bargiela A., Pedrycz W., Optimised information abstraction in granular Min/Max clustering, in S. Ramanna, R.J. Howlett, L. Jain (ed.), Emerging Paradigms in Machine Learning, Springer, June 2012. ISBN 978-3-642-28698-8
- Bargiela A., Pedrycz W., Supervised and unsupervised information granulation: A study in hyperbox design, Chapter 3 in J.T.Yao (ed.), Novel developments in Granular Computing: Applications for Advanced Human Reasoning and Soft Computation, IGI Global, 2010.
- Bargiela, A., Pedrycz W., Self-organising maps for interactive information granulation, in Neural Network Applications in IT, In : D. Wang, N.K.Lee (eds.), University of Malasia Press, 1-19, 2004
- Bargiela, A. Pedrycz, W., Combined physical and mathematical model of Granular Computing, Proc. Int. Conf. On Soft Computing and Intelligent Systems SCIS2006, Tokyo, Japan, Sept. 2006