Fusion of anatomical and functional images using parallel saliency features

An efficient method is proposed for fusion of anatomical and functional images by constructing the fused image through the combination of parallel saliency features in a multi-scale domain. First, the anatomical and functional images are decomposed into a series of smooth layers and detail layers at different scales by the average filter. Second, the parallel saliency features of both sharp edge and color detail are extracted to obtain the saliency maps. The edge saliency weighted map aims to preserve the high-spatial-resolution structural information using the Canny edge detection operator, while the color saliency weighted map extracts the high-intensity color detail using the context-aware operator. Finally, the fused image is reconstructed by the fused smooth layers and the fused detail layers using saliency maps. We demonstrate the application of the proposed method to a medical problem: Alzheimer's disease. Experimental results show that the proposed method for fusion of MRI-CBV and SPECT-Tc images and fusion of MRI-T1 and PET-FDG images successfully presents the pleasing fused medical images with high-spatial-resolution anatomical structural boundaries and high-intensity color detail.