ABSTRACT
This chapter aims to provide an overview of several related approaches to the analysis of nonlinear processes that occur in vision. The term “systems analysis” denotes a theoretical framework for the characterization of the input-output properties of physical systems. To place nonlinear systems analysis in context, the chapter highlights some important ideas from the analysis of linear systems. The superposition property leads to an exhaustive but concise characterization of the input-output behavior of a linear system. The chapter considers a system with a single input, and discusses how the ideas extend to systems with two or more spatially distributed inputs. The Wiener representation of a nonlinear system is an alternative to the Volterra representation. In contrast to the Volterra kernels, the Wiener kernels are defined in terms of the response of the system to large signals, rather than to limitingly small ones.
