On the Design of Adaptive Image Filters with Type-2 Fuzzy Logics for Removal of Impulse Noises

• Main Authors: Yao-Ju Lee, 李耀儒
• Other Authors: Pao-Ta Yu
• Format: Others
• Language: en_US
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/08897401798721989523

碩士 === 國立中正大學 === 通訊工程研究所 === 93 === In this thesis, we present an image filter system based on type-2 Fuzzy Logic System (FLS). The type-1 fuzzy system was proposed by L. A. Zadeh in 1965 and has been widely applied in various fields such as digital signal processing, digital signal communications, decision-making concerning optimization problem and artificial intelligent. The type-2 fuzzy sets are proposed to deal with uncertain information. Therefore, the type-2 FLS can handle uncertainties by enhancing the degrees of freedom. A kernel component of a fuzzy inference system is knowledge base consisting of fuzzy IF-THEN rules. The membership function of a type-2 fuzzy set is tree-dimensional that provides new design degrees of freedom for handling uncertainties. In rules-based FLSs, the type-2 fuzzy sets let us model and minimize the effects of uncertainties. For experimental results, the type-2 fuzzy filter can improve the performance of the type-1 fuzzy filter on the removal of impulse noises.