Python+OpenCV 图像边缘检测四种实现方法

目录

• 1.sobel算子
• 2.schaar算子(更能体现细节)
• 3.laplacian算子(基于零穿越的，二阶导数的0值点)
• 4.canny边缘检测(被认为是最优的边缘检测算法)
• 总结

import cv2 as cv
import numpy as np
import matplotlib.pyplot as plt
# 设置兼容中文
plt.rcparams['font.family'] = ['sans-serif']
plt.rcparams['font.sans-serif'] = ['simhei']

d:\anaconda\azwz\lib\site-packages\numpy\_distributor_init.py:30: userwarning: loaded more than 1 dll from .libs:
d:\anaconda\azwz\lib\site-packages\numpy\.libs\libopenblas.noijjg62emaszi6nyurl6jbkm4evbgm7.gfortran-win_amd64.dll
d:\anaconda\azwz\lib\site-packages\numpy\.libs\libopenblas.wcdjnk7yvmpzq2me2zzhjjrj3jikndb7.gfortran-win_amd64.dll
  warnings.warn("loaded more than 1 dll from .libs:\n%s" %

horse = cv.imread('img/horse.jpg',0)

plt.imshow(horse,cmap=plt.cm.gray)

plt.imshow(horse,cmap=plt.cm.gray)

1.sobel算子

# 1,0 代表沿x方向做sobel算子
x = cv.sobel(horse,cv.cv_16s,1,0)
# 0,1 代表沿y方向做sobel算子
y = cv.sobel(horse,cv.cv_16s,0,1)

# 格式转换
absx = cv.convertscaleabs(x)
absy = cv.convertscaleabs(y)

# 边缘检测结果
res = cv.addweighted(absx,0.5,absy,0.5,0)

plt.figure(figsize=(20,20))
plt.subplot(1,2,1)
m1 = plt.imshow(horse,cmap=plt.cm.gray)
plt.title("原图")
plt.subplot(1,2,2)
m2 = plt.imshow(res,cmap=plt.cm.gray)
plt.title("sobel算子边缘检测")

text(0.5, 1.0, 'sobel算子边缘检测')

2.schaar算子(更能体现细节)

# 1,0 代表沿x方向做sobel算子
x = cv.sobel(horse,cv.cv_16s,1,0,ksize=-1)
# 0,1 代表沿y方向做sobel算子
y = cv.sobel(horse,cv.cv_16s,0,1,ksize=-1)

# 格式转换
absx = cv.convertscaleabs(x)
absy = cv.convertscaleabs(y)

# 边缘检测结果
res = cv.addweighted(absx,0.5,absy,0.5,0)

plt.figure(figsize=(20,20))
plt.subplot(1,2,1)
m1 = plt.imshow(horse,cmap=plt.cm.gray)
plt.title("原图")
plt.subplot(1,2,2)
m2 = plt.imshow(res,cmap=plt.cm.gray)
plt.title("schaar算子边缘检测")

text(0.5, 1.0, 'schaar算子边缘检测')

3.laplacian算子(基于零穿越的，二阶导数的0值点)

res = cv.laplacian(horse,cv.cv_16s)

res = cv.convertscaleabs(res)

plt.figure(figsize=(20,20))
plt.subplot(1,2,1)
m1 = plt.imshow(horse,cmap=plt.cm.gray)
plt.title("原图")
plt.subplot(1,2,2)
m2 = plt.imshow(res,cmap=plt.cm.gray)
plt.title("laplacian算子边缘检测")

text(0.5, 1.0, 'laplacian算子边缘检测')

4.canny边缘检测(被认为是最优的边缘检测算法)

res = cv.canny(horse,0,100)

# res = cv.convertscaleabs(res) canny边缘检测是一种二值检测，不需要转换格式这一个步骤

plt.figure(figsize=(20,20))
plt.subplot(1,2,1)
m1 = plt.imshow(horse,cmap=plt.cm.gray)
plt.title("原图")
plt.subplot(1,2,2)
m2 = plt.imshow(res,cmap=plt.cm.gray)
plt.title("canny边缘检测")

text(0.5, 1.0, 'canny边缘检测')

总结

以上就是python+opencv 图像边缘检测四种实现方法的详细内容，更多关于python opencv图像边缘检测的资料请关注www.887551.com其它相关文章！