Scientific computing toolkit SciPy image processing

Reading guide

scipy It also provides many methods about image processing , Mainly packaged in ndimage In bag , This article mainly introduces the use of some image processing methods

Image filtering

• Get photo

from scipy import misc,ndimagefrom matplotlib import pyplot as plt# Get a picture from the website img = misc.ascent()# Displays a grayscale image plt.gray()# display picture plt.imshow(f)plt.show()

• wave filtering
  wave filtering It is usually used for image denoising ,scipy Many filtering algorithms are also provided in

# Gauss filtering ,sigma Set the standard deviation of Gaussian kernel gauss_f_img = ndimage.gaussian_filter(img,sigma=3)# Maximum filtering max_f_img = ndimage.maximum_filter(img,size=20)# median filtering median_f_img = ndimage.median_filter(img,size=20)# Minimum filtering min_f_img = ndimage.minimum_filter(img,size=20)# Percentile filtering percentile_f_img = ndimage.percentile_filter(img,percentile=20,size=20)# Set the display of the picture listimg_list = [img,gauss_f_img,max_f_img,median_f_img,min_f_img,percentile_f_img]title_list = ["origin","guassian filter","maximum filter","median filter","minimum filter","percentile filter"]for i in range(len(img_list)):    plt.subplot(2,3,i+1)    # Set the title of the display picture     plt.title(title_list[i])    # display picture     plt.imshow(img_list[i])    # Turn off the display of the coordinate axis     plt.axis("off")plt.show()

Image interpolation

Image interpolation is also widely used in image processing , for example ： Perspective transformation 、 Affine transformation 、 translation 、 The zoom 、 rotate You can see it when you wait ,scipy Several functions are also provided for image interpolation , Let's take a look at

On the principle of image affine transformation , You can refer to my article Understand affine transformation

• Affine transformation

# Define the translation matrix of an image M = np.array([[1,0,10],[0,1,30]])# Affine transformation affine_img = ndimage.affine_transform(img,M)plt.imshow(affine_img)plt.show()

• Image rotation

# Rotated image #reshape Set whether to display all images ,True Show all ,False The image will be cropped rotate_img = ndimage.rotate(img,45,reshape=False)plt.imshow(rotate_img)plt.show()

• Image zooming

# Reduce the image to the original 1/2zoom_img = ndimage.zoom(img,0.5)plt.imshow(zoom_img)plt.show()

• Geometric interpolation
  adopt geometric_transform We can realize geometric interpolation , You only need to define a function of image position transformation , As defined below shift_func Function to move the image to the lower left corner 1 Pixel , The cross-border position uses... By default 0 Fill in

a = np.arange(0,12).reshape((4,3))def shift_func(output_coords):    # Define the interpolation function     return (output_coords[0] - 1,output_coords[1] - 1)# The pixel values of the coordinates are transformed according to the coordinates of the data t_a = ndimage.geometric_transform(a,shift_func)"""[[ 0  1  2] [ 3  4  5] [ 6  7  8] [ 9 10 11]][[0 0 0] [0 0 1] [0 3 4] [0 6 7]]"""