In this chapter, we will discuss mathematical morphology and morphological image processing. Morphological image processing is a collection of non-linear operations related to the shape or morphology of features in an image. These operations are particularly suited to the processing of binary images (where pixels are represented as 0 or 1 and, by convention, the foreground of the object = 1 or white and the background = 0 or black), although it can be extended to grayscale images.

In morphological operations, a structuring element (a small template image) is used to probe the input image. The algorithms work by positioning the structuring element at all possible locations in the input image and comparing it with the corresponding neighborhood of the pixels with a set operator. Some operations test whether the element fits within the neighborhood, while others test whether it hits or intersects the neighborhood. A few popular morphological operators or filters are binary dilation and erosion, opening and closing, thinning, skeletonizing, morphological edge detectors, hit or miss filters, rank filters, median filters, and majority filters.

This chapter will demonstrate how to use morphological operators or filters on binary and grayscale images along with their applications, using functions from the scikit-image and SciPy ndimage.morphology module. 

The topics to be covered in this chapter are as follows:

• Morphological image processing with the scikit-image morphology module
• Morphological image processing with the scikit-image filter.rank module
• Morphological image processing with the scipy.ndimage.morphology module