Pengolahan Citra Biner

 Deteksi Tepi

import numpy as np import matplotlib.pyplot as plt   from skimage import filters from skimage.data import camera from skimage.util import compare_images     image = camera() edge_roberts = filters.roberts(image) edge_sobel = filters.sobel(image)   fig, axes = plt.subplots(ncols=2, sharex=True, sharey=True,                          figsize=(8, 4))   axes[0].imshow(edge_roberts, cmap=plt.cm.gray) axes[0].set_title('Roberts Edge Detection')   axes[1].imshow(edge_sobel, cmap=plt.cm.gray) axes[1].set_title('Sobel Edge Detection')   for ax in axes:     ax.axis('off')   plt.tight_layout() plt.show()

Active_contour

import numpy as np import matplotlib.pyplot as plt from skimage.color import rgb2gray from skimage import data from skimage.filters import gaussian from skimage.segmentation import active_contour img = data.text()   r = np.linspace(136, 50, 100) c = np.linspace(5, 424, 100) init = np.array([r, c]).T   snake = active_contour(gaussian(img, 1), init, boundary_condition='fixed',                        alpha=0.1, beta=1.0, w_line=-5, w_edge=0, gamma=0.1)   fig, ax = plt.subplots(figsize=(9, 5)) ax.imshow(img, cmap=plt.cm.gray) ax.plot(init[:, 1], init[:, 0], '--r', lw=3) ax.plot(snake[:, 1], snake[:, 0], '-b', lw=3) ax.set_xticks([]), ax.set_yticks([]) ax.axis([0, img.shape[1], img.shape[0], 0])   plt.show()

Gaussian Noise

import numpy as np import matplotlib.pyplot as plt import skimage import skimage.color as skic import skimage.filters as skif import skimage.data as skid import skimage.util as sku %matplotlib inline

def show(img):     fig, ax = plt.subplots(1, 1, figsize=(8, 8))     ax.imshow(img, cmap=plt.cm.gray)     ax.set_axis_off()     plt.show()

img = skic.rgb2gray(skid.astronaut()) show(img)

show(skif.gaussian(img, 5.))

sobimg = skif.sobel(img) show(sobimg)

from ipywidgets import widgets   @widgets.interact(x=(0.01, .2, .005)) def edge(x):     show(sobimg < x)

img = skimage.img_as_float(skid.astronaut())   # We take a portion of the image to show the details. img = img[50:200, 150:300]   # We add Gaussian noise. img_n = sku.random_noise(img) show(img_n)

img_r = skimage.restoration.denoise_tv_bregman(     img_n, 5.)   fig, (ax1, ax2, ax3) = plt.subplots(     1, 3, figsize=(12, 8))   ax1.imshow(img_n) ax1.set_title('With noise') ax1.set_axis_off()   ax2.imshow(img_r) ax2.set_title('Denoised') ax2.set_axis_off()   ax3.imshow(img) ax3.set_title('Original') ax3.set_axis_off()

Share on :

Tweet