Characteristic Analysis of Textile Materials Based on Pre-Judgment Mechanism

Abstract

Aiming at existing some problems, such as large calculation, not extensive category of defect recognition, and not accuracy of recognition result, this paper reanalyzes characteristic construction’s proceed of moving image by human vision, and proposes a new ‘super-fuzzy’ characteristic model of fabric based on pre-judgment mechanism. In this model, ‘super-fuzzy’ factor is introduced, and pre-judgment mechanism on coarseness set is constructed. So the characteristic can be modified with optimization according to actual condition of image. From experiments, results show that this characteristic model has such features as calculation is less, fabric images analysis is fast, fabric and defect category is extensive, pre-learning of mechanism doesn’t need, and better application is expected.