Multimodal medical image fusion is used to merge functional and structural information of the same body organ. Most of the multimodal image fusion algorithms are designed to fuse grayscale images that are produced by ...Multimodal medical image fusion is used to merge functional and structural information of the same body organ. Most of the multimodal image fusion algorithms are designed to fuse grayscale images that are produced by different imaging modalities. It is likely that problems of colour distortion and information loss will occur in fused image when source images are fused by using algorithms that are not originally designed to fuse colour images. These problems can be avoided by representing and processing source images as quaternion numbers. Quaternion representation of a colour pixel encodes information of its colour channels on the imaginary parts of a quaternion number and provides the advantage to processing colour information holistically as a vector field. In this paper, we proposed an image fusion algorithm based on Quaternion Principal Component Analysis (QPCA), to fuse multimodal colour medical images.Quaternion principal components are calculated by decomposing quaternion covariance matrix using Quaternion Eigenvalue Decomposition (QEVD). Fusion rule is designed, based on the fusion weights that are extracted from the highly influential principal component. To test the performance of the proposed algorithm,experiments have been performed on six image-sets of multimodal colour images of the brain. Experimental results are compared objectively with existing image fusion algorithms. Comparison results show that the proposed algorithm performed better than existing algorithms in fusing colour medical images.展开更多
文摘Multimodal medical image fusion is used to merge functional and structural information of the same body organ. Most of the multimodal image fusion algorithms are designed to fuse grayscale images that are produced by different imaging modalities. It is likely that problems of colour distortion and information loss will occur in fused image when source images are fused by using algorithms that are not originally designed to fuse colour images. These problems can be avoided by representing and processing source images as quaternion numbers. Quaternion representation of a colour pixel encodes information of its colour channels on the imaginary parts of a quaternion number and provides the advantage to processing colour information holistically as a vector field. In this paper, we proposed an image fusion algorithm based on Quaternion Principal Component Analysis (QPCA), to fuse multimodal colour medical images.Quaternion principal components are calculated by decomposing quaternion covariance matrix using Quaternion Eigenvalue Decomposition (QEVD). Fusion rule is designed, based on the fusion weights that are extracted from the highly influential principal component. To test the performance of the proposed algorithm,experiments have been performed on six image-sets of multimodal colour images of the brain. Experimental results are compared objectively with existing image fusion algorithms. Comparison results show that the proposed algorithm performed better than existing algorithms in fusing colour medical images.
