sobel算子边缘检测并细化

sobel算子：

import numpy as np
import cv2
import os

def sobel(img, threshold):
    G_x = np.array([[-1, 0, 1], [-2, 0, 2], [-1, 0, 1]])
    G_y = np.array([[-1, -2, -1], [0, 0, 0], [1, 2, 1]])
    rows = np.size(img, 0)
    columns = np.size(img, 1)
    mag = np.zeros(img.shape)
    for i in range(0, rows - 2):
        for j in range(0, columns - 2):
            v = sum(sum(G_x * img[i:i + 3, j:j + 3]))
            h = sum(sum(G_y * img[i:i + 3, j:j + 3]))
            mag[i + 1, j + 1] = np.sqrt((v ** 2) + (h ** 2))

    for p in range(0, rows):
        for q in range(0, columns):
            if mag[p, q] < threshold:
                mag[p, q] = 0
    return mag

path_color="G:\PointCloudUDA\inputt\PnpAda_release_data\mr_train\color_mask/"
files = os.listdir(path_color)
files.sort(key=lambda x: (x[:-4]))
for i in range(len(files)):
    print(i)

    cloud_dir = path_color + files[i]
    img = cv2.imread(cloud_dir, 0)  # read an image
    mag = sobel(img, 10)
    hh=np.where(mag==0.0,0.0,255.0)
    # plt.imshow(hh,cmap="gray")
    # plt.show()
    cv2.imwrite("G:\PointCloudUDA\inputt\PnpAda_release_data\ct_train\line/"+files[i],hh)

细化：

import cv2
from skimage import morphology
import numpy as np
import glob
import os
from matplotlib import  pyplot as plt

imageList = "G:\PointCloudUDA\inputt\PnpAda_release_data\ct_train\line/"    
img_file = "G:\PointCloudUDA\inputt\PnpAda_release_data\ct_train\line_xi/"
files = os.listdir(imageList)
files.sort(key=lambda x: (x[:-4]))

# 批量处理图片
for i in range(len(files)):
    print(i)
    #np.empty((256*256, 4))
    cloud_dir = imageList + files[i]
    img = cv2.imread(cloud_dir, 0)
    binary=np.array(img/255)
    skeleton0 = morphology.skeletonize(binary)
    
    #method2 
    # skel, distance = morphology.medial_axis(binary, return_distance=True)  # 图片细化（骨架提取）
    # skeleton0 = distance * skel
    
    skeleton = skeleton0.astype(np.uint8)*255
    cv2.imwrite(img_file+files[i], skeleton)   

细化后结果：