In an image, points of interest are positions where there might be edges, corners, or interesting objects. For example, in a landscape picture, points of interest can be located near a house or a person. Detecting points of interest is useful in image recognition, computer vision, or medical imaging.
In this recipe, we will find points of interest in an image with scikit-image. This will allow us to crop an image around the subject of the picture, even when this subject is not in the center of the image.
Let's import the packages:
>>> import numpy as np import matplotlib.pyplot as plt import skimage import skimage.feature as sf %matplotlib inline
We create a function to display a colored or grayscale image:
>>> def show(img, cmap=None): cmap = cmap or plt.cm.gray fig, ax = plt.subplots(1, 1, figsize=(8, 6)) ax.imshow(img, cmap=cmap) ax.set_axis_off() return ax
We load...