ABSTRACT
A geometric transformation converts an image by rearranging the location of the pixels. A couple of simple examples are a linear shift or a rotation. This chapter discusses the theory of geometric transformation functions and methods by which they can be achieved using Python functions. Some of the functions are provided by scipy.ndimage, while others are constructed from basic principles. The most basic geometric operations are to extract a sub image from a larger image or to create a larger image from a set of smaller images. The Window operator extracts a sub image from the original. Arrays in Python contain information as rows and columns; however, there are some objects in images that are circular and angular in nature. Processing this information is easier in polar coordinates, but programming languages prefer rectilinear coordinates. Thus, in these cases, it is useful to have functions that transform information between rectilinear and polar coordinates.
