Image segmentation is the partitioning of an image into distinct regions or categories that correspond to different objects or parts of objects. Each region contains pixels with similar attributes, and each pixel in an image is allocated to one of these categories. A good segmentation is typically one in which pixels in the same category have similar intensity values and form a connected region, whereas the neighboring pixels that are in different categories have dissimilar values. The goal of this is to simplify/change the representation of an image into something more meaningful and easier to analyze. 

If segmentation is done well, then all other stages in image analysis are made simpler. Hence, the quality and reliability of segmentation dictates whether an image analysis will be successful. But to partition an image into correct segments is often a very challenging problem.

Segmentation techniques can be either non-contextual (do not consider spatial relationships between features in an image and group pixels only with regard to some global attributes, for example, color/gray level) or contextual (additionally exploit spatial relationships; for example, group spatially close pixels with similar gray levels). In this chapter, we will discuss different segmentation techniques and demonstrate their Python-based implementations using scikit-image, python-opencv (cv2) and SimpleITK library functions. Let's start by importing the required libraries for this chapter:

import numpy as np
from skimage.transform import (hough_line, hough_line_peaks, hough_circle, hough_circle_peaks)
from skimage.draw import circle_perimeter
from skimage.feature import canny
from skimage.data import astronaut
from skimage.io import imread, imsave
from skimage.color import rgb2gray, gray2rgb, label2rgb
from skimage import img_as_float
from skimage.morphology import skeletonize
from skimage import data, img_as_float
import matplotlib.pyplot as pylab
from matplotlib import cm
from skimage.filters import sobel, threshold_otsu
from skimage.feature import canny
from skimage.segmentation import felzenszwalb, slic, quickshift, watershed
from skimage.segmentation import mark_boundaries, find_boundaries