This paper proposes a model of perception that allows animals to classify objects in the environment. We consider the transformation of semantic information in the four blocks of the model that imitate the mechanism of operation of sensory systems. Receptors convert external influences into stimuli that are transformed into sensations in accordance with the law of requisite variety via the data randomizing. Perception is formed by the generalization of all sensations, as well as the corresponding information accumulated by the animal. To find appropriate prototypes of objects forming classes, there is a compression of the processed information. This process is modeled via the creation of granules containing objects with close values for every feature. Granulation allows us to find the most probable class of the object corresponding to the average frequency value of its features. Algorithms for object classification on basis of the model and the invariant paradigm are identical and posses the simplicity and versatility at a high accuracy of the solution.
CITATION STYLE
Shats, V. N. (2018). The Classification of Objects Based on a Model of Perception. In Studies in Computational Intelligence (Vol. 736, pp. 125–131). Springer Verlag. https://doi.org/10.1007/978-3-319-66604-4_19
Mendeley helps you to discover research relevant for your work.