Spectral and Textural Weighting Using Takagi-Sugeno Fuzzy System for Through Wall Image Enhancement

Abstract

Abstract-A through wall image enhancement scheme based on Takagi-Sugeno fuzzy system and principal component analysis is proposed. The scheme incorporates spectral properties of image and textural properties of eigen components of image to assign weights. The scheme overcomes the empirical setting of inference engine and output membership functions. Simulation demonstrates the effectiveness of proposed scheme in terms of accuracy.

Mri Brain Classification Using Texture Features, Fuzzy Weighting and Support Vector Machine

“…where is the t-norm operator, taken as the algebraic product, and p 1 and p 2 are positive parameters and used for noise suppression in input data, e.g., if p 1 and p 2 are larger than σ 2 [27,28]. Product Inference Engine (PIE) is used to process fuzzy inputs based on fuzzy rule base and linguistic rules [27,28].…”

“…Product Inference Engine (PIE) is used to process fuzzy inputs based on fuzzy rule base and linguistic rules [27,28]. Fuzzy IF-THEN rules for weight assignment are, Ru (1) : IF α is Low and β l is Low THEN w l is Very High.…”

“…Weights, calculated using fuzzy logic, are assigned to features based on its discrimination capability. Fuzzy logic is used for weight assignment as it provide better performance in case of overlapping boundaries of data [27,28]. Simulation results show that the proposed technique is able to classify between different abnormal classes and provide better classification accuracy compared to the existing state of art techniques.…”