A new approach of multi-modal medical image fusion using intuitionistic fuzzy set. (August 2022)

• Record Type: Journal Article
• Authors: Palanisami, Dhanalakshmi; Mohan, Nandhini; Ganeshkumar, Lavanya
• Is Part Of: Biomedical signal processing and control. Volume 77(2022)
• Publication Date: 2022-08
• Subjects: Multi-modal medical image fusion -- Two-scale decomposition -- Gaussian filter -- Texture feature; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.103762
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Two scale decomposition helps to extract the significant features of the image. Sugeno intutionistic fuzzy set effectively eradicates the uncertainty in the images. The proposed fusion algorithm provides the enhanced fusion result of grayscale medical images with good contrast and high structural information. The objective and subjective assessment shows the proposed fusion algorithm is superior than other methods. Abstract: The main aim of this paper is to blend the multi-modality images into a single image to acquire superior information and obtain excellent visual quality without any artifacts and vagueness. First, the source images are decomposed into base and detail layers by making use of the Gaussian filter. The detail layers are merged using spatial frequency to preserve the edge details and clarity of an image. Due to the reason that the base layer contains approximate information of source image with low contrast, it is transformed into Sugeno's intuitionistic fuzzy image (SIFI). Texture information is extracted from SIFI to fuse the base layers. Finally, the fused output image with enhanced contrast and better visual effects is reconstructed by integrating both the fused base and fused detail layer. Furthermore, the experimental results demonstrate the efficacy of the proposed method over other fusion methods. Both subjective and objective assessment illustrates that the proposed method presents a fused output image that has good visual quality with enhanced contrast and no artifacts included.