from scipy.misc import imresize
import graphcut
from scipy.cluster.vq import *
import sift
from numpy import *
from matplotlib.pyplot import *
from PIL import Image
im = array(Image.open("building.png"))[:,:,:3]
im = imresize(im,0.1,interp="bilinear")
size = im.shape[:2]
# add two rectangular training regions
labels = zeros(size)
labels[3:18,3:18] = -1
labels[-18:-3,-18:-3] = 1
# create graph
g = graphcut.build_bayes_graph(im,labels,kappa=1)
# cut the graph
res = graphcut.cut_graph(g,size)
figure()
graphcut.show_labeling(im,labels)
figure()
imshow(res)
gray()
axis("off")
show()
from scipy.misc import imresize
import graphcut
from PIL import Image
from numpy import *
from pylab import *

im = array(Image.open('../pcv_data/data/empire.jpg'))
im = imresize(im, 0.03, interp='bilinear')
size = im.shape[:2]

# add two rectangular training regions
labels = zeros(size)
labels[3:8, 3:8] = -1
labels[-8:-3, -8:-3] = 1

# create graph
g = graphcut.build_bayes_graph(im, labels, kappa=1)

# cut the graph
res = graphcut.cut_graph(g, size)

figure()
graphcut.show_labeling(im, labels)

figure()
imshow(res)
gray()
axis('off')

show()
Esempio n. 3
0
File: 0626.py Progetto: ta-oyama/PCV
size = im.shape[:2]

plt.imshow(im)

#2つの長方形の訓練データ領域を追加する
labels = np.zeros(size)
labels[3:18,3:18] = -1
labels[-18:-3,-18:-3] = 1

plt.imshow(labels)

#グラフを作成する
g = graphcut.build_bayes_graph(im,labels,kappa=1)

#グラフをカットする
res = graphcut.cut_graph(g,size)

plt.figure()
graphcut.show_labeling(im,labels)

plt.figure()
plt.imshow(res)
plt.gray()
plt.axis('off')


#9.1.2 ユーザー入力を用いた領域分割

#.tarを展開する
#import os
#os.chdir("C:\Users\minako\Downloads\images")
Esempio n. 4
0
# create graph
g = graphcut.build_bayes_graph(im, labels, kappa=1)


out = open("tocpp.txt", "w")
out.write(str(len(g)) + "\n")

edlist = g.edges()
for a in edlist:
    out.write(str(a[0]) + " " + str(a[1]) + " " + str(g.edge_weight((a[0], a[1]))) + "\n")
out.write(str(-1) + " " + str(-1) + " " + str(-1) + "\n")
out.close()


algo = r"edmond_karp.exe"
# algo=r"push_relabel.exe"
print algo

# cut the graph
res = graphcut.cut_graph(g, size, algo)


# figure()
# graphcut.show_labeling(im,labels)

figure()
imshow(res)
gray()
axis("off")
show()
9.1. Graph Cuts 245
pixel value
meaning
0, 64
background
128
unknown
255
foreground
                                            
      # load image and annotation map
im = array(Image.open(’376043.jpg’))
m = array(Image.open(’376043.bmp’))
# resize
scale = 0.1
im = imresize(im,scale,interp=’bilinear’)
m = imresize(m,scale,interp=’nearest’)
# create training labels
labels = create_msr_labels(m,False) # build graph using annotations
g = graphcut.build_bayes_graph(im,labels,kappa=2) # cut graph
res = graphcut.cut_graph(g,im.shape[:2])
# remove parts in background
res[m==0] = 1
res[m==64] = 1
# plot the result
figure()
imshow(res)
gray()
xticks([])
yticks([])
savefig(’labelplot.pdf’)