Medical image fusion technique based on type-2 near fuzzy set

This paper presents a fuzzy fusion technique for multimodal medical image fusion using type-2 fuzzy set and near set. For each pixel of both source images, pixel-wise fuzzification based on histogram level is done. Then, construct type-2 fuzzy membership grade from both fuzzified images to quantify the uncertainty of shape of membership function. Later, fuzzy entropy, mutual information and correlation are estimated the fuzziness, correlated fuzzy information, and similarity between both source images. Near fuzzy set is used to estimate the amount of nearness between two membership grades and combined best membership grade. Finally, fused image is obtained by defuzzification. The proposed method is simulated on a set of medical images and compared with state-of-the-art methods. The experimental results demonstrates the superiority of the proposed technique in terms of various fusion metrics.