Image Fusion is the procedure of combining useful features from multiple sensor image inputs to form a single composite image. In this work, the authors extend the previously proposed Image Fusion framework, based on self-trained Independent Component Analysis (ICA) bases, to a more sophisticated region-based Image Fusion system. The input images are segmented into three areas of different activity : edges, texture and constant background. A hierarchical set of fusion rules employing textural information from the spatial-domain in the form of local variance, entropy and fourier energy is introduced. The proposed system improves the performance of our previous system.