A Medical Image Fusion Method Based on Visual Models
A new method of medical image fusion is proposed in this paper, which is based on human visual models and IHS color space. Retina-inspired difference of Gaussian model is adopted to enhance the spatial information of anatomical images. Also, 2D Log-Gabor model of primary visual cortex is used to enhance the spectrum information of functional images. The statistical analyses tools such as average gradient and entropy are demonstrated that the proposed algorithm does considerably increase spatial information content and reduce the color distortion compared to the counterpart fusion methods. In the proposed fused images the color information is least distorted, the spatial details are as clear as the original anatomical images, and the integration of color and spatial features was normal.
Keywordsimage fusion visual models IHS color space difference of Gaussian model 2D Log-Gabor model
Unable to display preview. Download preview PDF.
- 5.Zhang, H., Liu, L., Lin, N.: A novel wavelet medical image fusion method. In: International Conference on Multimedia and Ubiquitous Engineering, pp. 548–553 (April 2007)Google Scholar
- 8.Zhao, W., Huang, J.J., Tian, B.: An image fusion algorithm based on receptive field model. Acta Electronica Sinica 36(9), 1665–1669 (2008)Google Scholar
- 10.Ghassemian, H.: A retina based multi-resolution image fusion. In: IGRSS 2001, pp. 709–711 (July 2001)Google Scholar