Optimum spectrum mask based medical image fusion using Gray Wolf Optimization. (April 2017)

• Record Type: Journal Article
• Authors: Daniel, Ebenezer; Anitha, J.; Kamaleshwaran, K.K; Rani, Indu
• Is Part Of: Biomedical signal processing and control. Volume 34(2017)
• Page Start: 36
• Page End: 43
• Publication Date: 2017-04
• Subjects: Image fusion -- Spectrum mask -- Gray Wolf Optimization -- Unsharp masking -- Multi scaling; 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.2017.01.003
• Languages: English
• ISSNs: 1746-8094

Abstract

Graphical abstract: Highlights: Proposed a new method, optimum spectrum mask fusion for multimodal medical image fusion. Gray wolf optimization is introduced for scale values selection in multi modal medical image fusion. More information of image and edge are transformed to the fused image using optimum masking. The proposed algorithm performed better results compared to the existing multi modal fusion techniques. Abstract: Medical image fusion techniques have been extensively used in clinical diagnosis. Scaling based techniques are well known techniques in multimodal image fusion, the generalized scaling has static scale value selection which brims the quality of fusion. In this paper, we proposed an Optimum Spectrum Mask Fusion (OSMF) for medical image fusion using conventional Gray Wolves Optimization (GWO) algorithm. The GWO algorithm, indulges swiftest and dynamic scale selection. The spectrum masks technique, contrary to conventional spatial domain and transform domain fusion algorithms in terms of image contrast and edge quality. The optimum mask caters more information in the multi-modality fusion. The proposed OSMF is tested for MR-SPECT, MR-PET, MR-CT and MR: T1-T2 of brain images. Experimental results show, our technique pageantries the improved results than other conventional pixel based fusion techniques.