Medical image fusion facilitates the retrieval of complementary information from medical images and has been employed diversely for computer-aided diagnosis of life threatening diseases. Fusion has been performed using various approaches such as Pyramidal, Multi-resolution, multi-scale etc. Each and every approach of fusion depicts only a particular feature (i.e. the information content or the structural properties of an image). Therefore, this paper presents a comparative analysis and evaluation of multi-modal medical image fusion methodologies employing wavelet as a multi-resolution approach and ridgelet as a multi-scale approach. The current work tends to highlight upon the utility of these approaches according to the requirement of features in the fused image. Principal Component Analysis (PCA) based fusion algorithm has been employed in both ridgelet and wavelet domains for purpose of minimisation of redundancies. Simulations have been performed for different sets of MR and CT-scan images taken from ‘The Whole Brain Atlas'. The performance evaluation has been carried out using different parameters of image quality evaluation like: Entropy (E), Fusion Factor (FF), Structural Similarity Index (SSIM) and Edge Strength (QFAB). The outcome of this analysis highlights the trade-off between the retrieval of information content and the morphological details in finally fused image in wavelet and ridgelet domains.
Different objects could be analyzed by use of pyramid decomposition of image. Image fusion algorithm based on pyramid decomposition of image is multi-scale, multi-resolution method. Its process is completed on different scale, different resolution and different decomposition layer. Compared with spatial fusion method, fusion effects improve obviously. In this paper, principal of pyramid decomposition and process were introduced, and simulation results of image fusion based on Laplacian pyramid, gradient pyramid, ration pyramid and contrast shows that image fusion based on pyramid decomposition is improve obviously.
An Enhanced Image Fusion Algorithm by Combined Histogram Equalization and Fast Gray Level Grouping Using Multi-Scale Decomposition and Gray-PCA
