Abstract
The fundamentals of object orientation are studied in the context of fuzzy sets. The main concepts such as inheritance are carefully analyzed in this setting with an intent to take full advantage of a linguistic generalization while retaining the key features of the object-oriented approach. As the admitted values of the attributes are linguistic, the proposed method hinges on a concept of linguistic approximation and computational enhancements stemming from the theory fuzzy neural networks. A numerical illustration comes from the realm of scene analysis and interpretation.