[人工智能]数字图像处理 使用skimage进行图像分割

基于距离变换的分水岭算法

代码如下：

import cv2 as cv2
import numpy as np
from skimage import morphology
from skimage.color import label2rgb
from skimage import feature
from skimage import segmentation


#读取图像，支持 bmp、jpg、png、tiff 等常用格式
def imregionalmax(image, ksize=3):
  """Similar to matlab's imregionalmax"""
  filterkernel = np.ones((ksize, ksize)) # 8-connectivity
  reg_max_loc = feature.peak_local_max(image,
                               footprint=filterkernel, indices=False,
                               exclude_border=0)
  return reg_max_loc.astype(np.uint8)

def imcomplement(image):
  """Equivalent to matlabs imcomplement function"""
  min_type_val = np.iinfo(image.dtype).min
  max_type_val = np.iinfo(image.dtype).max
  return min_type_val + max_type_val - image


# 读取图像
image = cv2.imread('pears.png', cv2.IMREAD_GRAYSCALE)
sobelx = cv2.Sobel(image, cv2.CV_64F, 1, 0, ksize=3)
sobely = cv2.Sobel(image, cv2.CV_64F, 0, 1, ksize=3)
# 计算梯度幅度
grad_magn = np.sqrt(sobelx**2 + sobely**2)
# 拉伸到0-255范围
grad_magn = 255*(grad_magn - np.min(grad_magn)) / (np.max(grad_magn) - np.min(grad_magn))

selem = morphology.disk(20)
opened = morphology.opening(image, selem)
eroded = morphology.erosion(image, selem)
opening_recon = morphology.reconstruction(seed=eroded, mask=image, method='dilation')
closed_opening = morphology.closing(opened, selem)

dilated_recon_dilation = morphology.dilation(opening_recon, selem)
recon_erosion_recon_dilation = morphology.reconstruction(dilated_recon_dilation,
                                                    opening_recon,
                                                    method='erosion').astype(np.uint8)

# recon_dilation_recon_dilation = morphology.reconstruction(imcomplement(dilated_recon_dilation),
#                                                      imcomplement(opening_recon),
#                                                      method='dilation').astype(np.uint8)
# recon_dilation_recon_dilation_c = imcomplement(recon_dilation_recon_dilation)


foreground_1 = imregionalmax(recon_erosion_recon_dilation, ksize=65)
fg_superimposed_1 = image.copy()
fg_superimposed_1[foreground_1 == 1] = 255
foreground_2 = morphology.closing(foreground_1, np.ones((5, 5)))
foreground_3 = morphology.erosion(foreground_2, np.ones((5, 5)))
foreground_4 = morphology.remove_small_objects(foreground_3.astype(bool), min_size=20)

_, labeled_fg = cv2.connectedComponents(foreground_4.astype(np.uint8))
col_labeled_fg = label2rgb(labeled_fg)

_, thresholded = cv2.threshold(recon_erosion_recon_dilation, 0, 255, cv2.THRESH_BINARY+cv2.THRESH_OTSU)

#基于距离变换的分水岭算法
labels = morphology.watershed(grad_magn, labeled_fg)
superimposed = image.copy()
watershed_boundaries = segmentation.find_boundaries(labels)
superimposed[watershed_boundaries] = 255
superimposed[foreground_4] = 255
col_labels = label2rgb(labels)
col_labels_merged = label2rgb(labels, image)

labels = np.array(labels, dtype=np.uint8)
cv2.namedWindow("Image")
cv2.imshow("Image",labels * 10)
cv2.waitKey(0)

?

?