图像分割

from scipy.misc import imresize from PCV.tools import graphcut from PIL import Image from pylab import * def create_msr_labels(m, lasso=False): """ Create label matrix for training from user annotations. """ labels = zeros(im.shape[:2]) # background labels[m == 0] = -1 labels[m == 64] = -1 # foreground if lasso: labels[m == 255] = 1 else: labels[m == 128] = 1 return labels # 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')