A new MRI and PET image fusion algorithm based on pulse coupled neural network

The fusion of multimodal brain imaging for a given clinical application is a very important performance. Generally, the PET (positron emission tomography) image indicates the brain function and it has a low spatial resolution, the MRI image shows the brain tissue anatomy and contains no functional information. In this paper, we propose a novel medical image fusion algorithm which enhances the spatial resolution of the functional images by combining them with a high-resolution anatomic image. In the event, after the registration process, perform YCbCr on the multispectral image and get luminance, blue-difference and red difference chromatic components, and then DWT (discrete wavelet transform) image fusion algorithm based on PCNN (pulse coupled neural networks) is applied to fuse the MRI image and the luminance component (Y). Ultimately, fused image is obtained by inverse YCbCr transform of the new luminance and the old blue-difference and red difference chromatic components back into RGB space. An important feature of the algorithm is to use PCNN because it has the global couple and pulse synchronization characteristics. It has been proven suitable for image processing and successfully employed in image fusion. Our approach is compared with YCbCr, DWT, Contourlet, Curvelet methods. Results show proposed method preserves more spectral features with less spatial distortion.