ABSTRACT
The field of medical image fusion is faced with the problems of veracity, velocity, and volume of the data that require faster and efficient processing of information. This chapter provides an overview of information fusion techniques making use of feature and data fusion principles that find application in medical image computing and analysis. Feature-level fusion between images is challenged by the problem of interimage variability such as pixel mismatches, missing pixels, image noise, resolution, and contrast. Morphological operators make use of the connectedness between pixels either to improve the spatial arrange of the pixels or to distort them to extract useful features from the subset of spatially localized pixel features. Wavelet transforms have the ability to compress the details of the images through their coefficients and to separate the fine and coarse details from one another. Artificial neural networks (ANNs) represent a set of decision processing models inspired from the working of the human neural network.
