Python 图像处理(八) 基于卷积操作的图像增强

基于卷积操作的图像增强

1. 基于低通滤波的图像去噪

from PIL import Image, ImageFilter
import matplotlib.pylab as plt
import numpy as np

i = 1
plt.figure(figsize=(10,35))
for prop_noise in np.linspace(0.05,0.3,6):
    im = Image.open('images/mandrill.jpg')
    #print(np.max(im))
    # choose 5000 random locations inside image
    n = int(im.width * im.height * prop_noise)
    x, y = np.random.randint(0, im.width, n), np.random.randint(0, im.height, n)
    for (x,y) in zip(x,y):
        im.putpixel((x, y), ((0,0,0) if np.random.rand() < 0.5 else (255,255,255))) # geenrate salt-and-pepper noise
    #im.save('../images/mandrill_spnoise_' + str(prop_noise) + '.jpg')
    plt.subplot(6,2,i)
    plt.imshow(im)
    plt.title('Original Image with ' + str(int(100*prop_noise)) + '% added noise', size=15)
    i += 1
    im1 = im.filter(ImageFilter.BLUR)
    plt.subplot(6,2,i)
    plt.imshow(im1)
    plt.title('Blurred Image', size=20)
    i += 1
im.show()

plt.figure(figsize=(20,7))
plt.subplot(1,3,1)
plt.imshow(im)
plt.title('Original Image', size=30)
plt.axis('off')
for n in [3,5]:
    box_blur_kernel = np.reshape(np.ones(n*n),(n,n)) / (n*n)
    #print(box_blur_kernel)
    im1 = im.filter(ImageFilter.Kernel((n,n), box_blur_kernel.flatten()))
    plt.subplot(1,3,(2 if n==3 else 3))
    plt.imshow(im1)
    plt.title('Blurred with kernel size = ' + str(n) + 'x' + str(n), size=20)
    plt.axis('off')
plt.suptitle('PIL Mean Filter (Box Blur) with different Kernel size', size=30)
plt.show()

2. 基于卷积操作的边界增强

from PIL import Image
from PIL import ImageFilter
from PIL.ImageFilter import (FIND_EDGES, EDGE_ENHANCE, EDGE_ENHANCE_MORE)
im = Image.open('images/building.jpg')
plt.figure(figsize=(18,25))
plt.subplot(221)
plt.imshow(im)
plt.title('original (UMBC library)', size=20)
plt.axis('off')
i = 2
for f in (FIND_EDGES, EDGE_ENHANCE, EDGE_ENHANCE_MORE):
    plt.subplot(2,2,i)
    im1 = im.filter(f)
    plt.imshow(im1)
    plt.title(str(f), size=20)
    plt.axis('off')
    i += 1
plt.show()