Scipy ndimage (n- dimensional image) is a widely used module for image processing.
Below are among the most common tasks performed in image processing using ndimage:
Input, outputting, and displaying images.
Cropping an image, inverting an image, rotating an image, etc.
Image filtering operations such as denoising or sharpening an image.
Image Classification.
Feature extraction.
from scipy import miscimport matplotlib.pyplot as pltimg = misc.face()plt.imshow(img)plt.show()## lets look at some statistics like, mean, max, min valuesprint(img.mean(), img.max(), img.min())
All images in Python are numbers ranging between 0 and 255. Lets perform
import matplotlib.pyplot as pltimport numpy as npimport scipy.ndimage as ndimagefrom scipy import miscimg = misc.face()# Cropping an imagex = img.shape[0]y = img.shape[1]cropped = img[int(x / 4): int(- x / 4), int(y / 4): int(- y / 4)]plt.imshow(cropped)plt.show()plt.savefig('output/1.png')# inverted imageinverted = np.flipud(img)plt.imshow(inverted)plt.show()plt.savefig('output/2.png')# rotationrotating_by_45_degrees = ndimage.rotate(img, 45)plt.imshow(rotating_by_45_degrees)plt.show()plt.savefig('output/3.png')# blurring using gaussian filterblurred = ndimage.gaussian_filter(img, sigma=8)plt.imshow(blurred)plt.show()plt.savefig('output/4.png')
Free Resources