def wavelet_subtraction(img, level):
"""6- 7 level is recommended"""
if level == 0:
return img
wp = WaveletPacket2D(data=img, wavelet='haar', mode='sym')
back = resize(np.array(wp['a'*level].data), img.shape, order=3, mode='reflect')/(2**level)
img = img - back
return img
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import numpy as np
import matplotlib.pyplot as plt
from PIL import Image
from pywt import WaveletPacket2D
im = Image.open("data/aero.png").convert('L')
arr = np.fromstring(im.tostring(), np.uint8)
arr.shape = (im.size[1], im.size[0])
wp2 = WaveletPacket2D(arr, 'db2', 'sym', maxlevel=2)
# Show original figure
plt.imshow(arr, interpolation="nearest", cmap=plt.cm.gray)
path = ['d', 'v', 'h', 'a']
# Show level 1 nodes
fig = plt.figure()
for i, p2 in enumerate(path):
ax = fig.add_subplot(2, 2, i + 1)
ax.imshow(np.sqrt(np.abs(wp2[p2].data)),
origin='image',
interpolation="nearest",
cmap=plt.cm.gray)
ax.set_title(p2)
# Show level 2 nodes
import numpy as np
from sklearn.cluster import KMeans
#import matplotlib.pyplot as plt
from pywt import WaveletPacket2D
import cv2
Num_Img = 5000
DIR = './titles/'
pos = './pos_examples'
neg = './neg_examples'
file_list = os.listdir(DIR)
samples = []
for i in range(Num_Img):
gray = cv2.imread(DIR + file_list[i], 0)
wp2 = WaveletPacket2D(gray, 'db2', 'symmetric', maxlevel=1)
print(file_list[i])
path = ['d', 'v', 'h', 'a']
feature = np.zeros([4, wp2['d'].data.size])
for j, p2 in enumerate(path):
feature[j, :] = np.reshape(wp2[p2].data, [1, wp2[p2].data.size])
f = np.cov(feature)
f = f.reshape(f.size)
samples.append(f)
samples = np.array(samples)
kmeans = KMeans(n_clusters=2, random_state=0).fit(samples)
l = kmeans.labels_
for i in range(Num_Img):
if l[i] == 1:
shutil.copy(DIR + file_list[i], pos)