Computer based diagnosis of Alzheimer's disease can be performed by dint of the analysis of the functional and structural changes in the brain. Multispectral image fusion deliberates upon fusion of the complementary information while discarding the surplus information to achieve a solitary image which encloses both spatial and spectral details. This paper presents a Non-Sub-sampled Contourlet Transform (NSCT) based multispectral image fusion model for computer-aided diagnosis of Alzheimer's disease. The proposed fusion methodology involves color transformation of the input multispectral image. The multispectral image in YIQ color space is decomposed using NSCT followed by dimensionality reduction using modified Principal Component Analysis algorithm on the low frequency coefficients. Further, the high frequency coefficients are enhanced using non-linear enhancement function. Two different fusion rules are then applied to the low-pass and high-pass sub-bands: Phase congruency is applied to low frequency coefficients and a combination of directive contrast and normalized Shannon entropy is applied to high frequency coefficients. The superiority of the fusion response is depicted by the comparisons made with the other state-of-the-art fusion approaches (in terms of various fusion metrics).

Multimodal medical image sensor fusion has revolutionized the medical analysis by improving the precision of computer assisted diagnosis. This is incorporated by highlighting the complementary information while minimizing the redundant content in the fused images from various biomedical sensors like MRI, Computed Tomography, and Positron Emission Tomography/Single-Photon Emission Computerized Tomography. Multispectral image fusion is a special case of multimodal fusion which serves to encompass both spatial and spectral details in the fused image. This paper presents a hybrid sub-band decomposition scheme for multispectral image fusion comprising of non-subsampled contourlet transform and shearlet transform domains. The pre-processing stage involves color transformation of an input multispectral image from red-green-blue to YIQ color space. Thereafter, both the source images (i.e., panchromatic and multispectral images) after sub-band decomposition are processed via the application of contrast enhancement, weighted-principal component analysis, and max-max algorithms. The low frequency coefficients are processed via phase congruency whereas a combination of directive contrast and normalized Shannon entropy is applied to high frequency coefficients. The objective assessment of image quality has been carried out using various reference and no-reference based performance metrics. The distinguishing fusion response of the proposed hybrid scheme has been validated by the comparisons done with the other fusion approaches.

No abstract

No abstract

No abstract