The purpose of the book is to advance in the understanding of brain functions by defining a general framework for representation based on granular computing theory. The idea is to bring this mathematical formalism into the domain of fuzzy neural representation of visual cognition, setting the basis for a theory of image understanding and learning of visual tasks. Our granular system is abstracted from brain science and information science, and it can be embedded in a neural dynamical system, it is helpful to guide designing of brain like computer for the tasks of image understanding. In a complicate task of visual cognition, dynamic cooperation and competition in the interacting procedure of different substances, neuron cells, neural system structures are necessary. We take the task of image matting as an example to demonstrate that our multi-scale GrC has high ability to increase the texture information entropy and improve the effect of object-background separating.