A novel method for resemblance images using near fuzzy set

摘要

We introduce a new method Near-Fuzzy set for analysis image. Indeed, near sets are considered a generalization of the rough sets theory. A set X is close to another set Y insofar as the description of at least one of the object of X corresponds to the description of least one of objects of Y. Find the tolerance classes with objects of the same description is a major problem. Maximal Clique Enumeration Algorithm solves the same problem and improves field performance image resemblance. We propose an innovative technique that hybrids both near sets approach with Fuzzy sets approaches. In this paper we use the Near-Fuzzy sets method to obtain better results in the resemblance of facial images. The performance of use of near set approach has been proved throughout the Japanese Female Facial Expression (JAFFE) database.