OpenCV 图像处理工具包

1. 简介
2. 安装
3. 导入 opencv
4. 使用介绍
5. 核心操作
6. 参考链接

1 简介

OpenCV 是一个基于 BSD 许可（开源）发行的跨平台计算机视觉库，可以运行在 Linux、 Windows、Android 和 Mac OS 操作系统上。它轻量级而且高效——由一系列 C 函数和少量 C++ 类构成，同时提供了 Python、Ruby、MATLAB 等语言的接口，实现了图像处理和计算机视觉方面的很多通用算法。

2 安装

建议使用 pip 来安装

pip install opencv-python

3 导入 opencv

import cv2 as cv
import numpy as np
lenna = 'static/image/2019/01/lenna.png'

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/tmp/babel-a6MpLG/python-Z9sKMV", line 1, in <module>
    import cv2 as cv
ModuleNotFoundError: No module named 'cv2'

4 使用介绍

4.1 图片读写

4.1.1 读取图片

opencv 的彩色的图片采用 BGR 的排列方式，和我们通常使用 RGB 的方式有所不同。

img = cv.imread(lenna)
print('img type = {}'.format(type(img)))
print('img.shape = {}'.format(img.shape))
gray_image = cv.imread(lenna, cv.IMREAD_GRAYSCALE)
print('gray_image.shape = {}'.format(gray_image.shape))

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/tmp/babel-a6MpLG/python-umPlTP", line 1, in <module>
    img = cv.imread(lenna)
NameError: name 'cv' is not defined

4.1.2 显示图片

cv.imshow('image', img)
cv.waitKey(0)
cv.destroyAllWindows()

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/tmp/babel-a6MpLG/python-7U80YK", line 1, in <module>
    cv.imshow('image', img)
NameError: name 'cv' is not defined

4.1.3 写入图片

使用 matplotlib 来存储图片

import cv2 as cv
import numpy as np
import matplotlib.pyplot as plt
filename = '../static/image/2019/01/image094201.png'
img = cv.imread('../static/image/2019/01/lenna.png')
img = cv.cvtColor(img, cv.COLOR_BGR2RGB)
plt.imshow(img)
plt.savefig(filename)
return filename

使用 opencv 来存储图片

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv
filename = '../static/image/2019/01/image094813.png'
img = cv.imread('../static/image/2019/01/lenna.png')
gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
cv.imwrite(filename, gray)
return filename

4.2 视频操作

4.2.1 读取摄像头数据

import numpy as np
import cv2 as cv
cap = cv.VideoCapture(0)
while(True):
  # Capture frame-by-frame
  ret, frame = cap.read()
  # get/set props
  # frame_width = cap.get(cv.CAP_PROP_FRAME_WIDTH)
  # frame_height = cap.get(cv.CAP_PROP_FRAME_HEIGHT)
  # ret = cap.set(cv.CAP_PROP_FRAME_WIDTH, 320)
  # ret = cap.set(cv.CAP_PROP_FRAME_HEIGHT, 240)
  # Our operations on the frame come here
  gray = cv.cvtColor(frame, cv.COLOR_BGR2GRAY)
  # Display the resulting frame
  cv.imshow('frame', gray)
  if cv.waitKey(1) & 0xFF == ord('q'):
    break
# When everything done, release the capture
cap.release()
cv.destroyAllWindows()

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/tmp/babel-a6MpLG/python-WYiwSF", line 2, in <module>
    import cv2 as cv
ModuleNotFoundError: No module named 'cv2'

4.2.2 播放视频文件

import numpy as np
import cv2 as cv
cap = cv.VideoCapture('vtest.avi')
while(cap.isOpened()):
  ret, frame = cap.read()
  gray = cv.cvtColor(frame, cv.COLOR_BGR2GRAY)
  cv.imshow('frame', gray)
  if cv.waitKey(1) & 0xFF == ord('q'):
    break
cap.release()
cv.destroyAllWindows()

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/tmp/babel-a6MpLG/python-KdHI49", line 2, in <module>
    import cv2 as cv
ModuleNotFoundError: No module named 'cv2'

4.2.3 保存视频文件

import numpy as np
import cv2 as cv
cap = cv.VideoCapture(0)
# Define the codec and create VideoWriter object
fourcc = cv.VideoWriter_fourcc(*'XVID')
out = cv.VideoWriter('output.avi', fourcc, 20.0, (640, 480))
while(cap.isOpened()):
  ret, frame = cap.read()
  if ret == True:
    frame = cv.flip(frame, 0)
  # write the flipped frame
  out.write(frame)
  cv.imshow('frame', frame)
  if cv.waitKey(1) & 0xFF == ord('q'):
    break
  else:
    break
# Release everything if job is finished
cap.release()
out.release()
cv.destroyAllWindows()

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/tmp/babel-a6MpLG/python-6ru6pI", line 2, in <module>
    import cv2 as cv
ModuleNotFoundError: No module named 'cv2'

4.3 画图函数

4.3.1 画几何图形

画几何图形，包括：线条、圆、矩形、椭圆等

import numpy as np
import cv2 as cv
import matplotlib.pyplot as plt
filename = '../static/image/2019/01/image100920.png'

# Create a black image
img = np.zeros((320, 512, 3), np.uint8)

# cv.line(img, pt1, pt2, color, thickness)
cv.line(img, (0, 0), (300, 100), (255, 0, 0), 5)

# cv.rectangle(img, pt1, pt2, color, thickness
cv.rectangle(img, (384, 0), (510, 128), (0, 255, 0), 3)

# cv.circle(img, center, radius, color, thickness)
cv.circle(img, (447, 63), 63, (0, 0, 255), -1)

# cv.ellipse(img, center, axes, angle, startAngle, endAngle, color, thickness)
cv.ellipse(img, (256, 256), (100, 50), 0, 0, 180, 255, -1)

pts = np.array([[10, 5], [20, 30], [70, 20], [50, 10]], np.int32)
pts = pts.reshape((-1, 1, 2))
cv.polylines(img, [pts], True, (0, 255, 255))

plt.imshow(img)
plt.savefig(filename)
# cv.imwrite(filename, img)
return filename

4.3.2 写文字

import numpy as np
import cv2 as cv
import matplotlib.pyplot as plt
filename = '../static/image/2019/01/image102219.png'
img = np.ones((320, 512, 3), np.uint8) * 255

# cv.putText(img, text, org, font, fontScale, color, thickness, bottonLeftOrigin)
img = cv.putText(img, 'OpenCV', (100, 100), cv.FONT_HERSHEY_PLAIN, 4, (0, 0, 255), 2, cv.LINE_AA)
# cv.imwrite(filename, img)
plt.imshow(img)
plt.savefig(filename)
return filename

5 核心操作

5.1 基本操作

5.1.1 像素操作

img = cv.imread(lenna)
print('px[4, 5] = {}'.format(img[4, 5]))
print('blut px[4, 5, 0] = {}'.format(img[4, 5, 0]))

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/tmp/babel-a6MpLG/python-E8wgao", line 1, in <module>
    img = cv.imread(lenna)
NameError: name 'cv' is not defined

5.1.2 相片属性

img = cv.imread(lenna)
print('shape = {}, size = {}, dtype = {}'.format(img.shape, img.size, img.dtype))

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/tmp/babel-a6MpLG/python-4waj0r", line 1, in <module>
    img = cv.imread(lenna)
NameError: name 'cv' is not defined

5.1.3 子图片和通道

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv
filename = '../static/image/2019/01/image135851.png'
lenna = '../static/image/2019/01/lenna.png'

img = cv.cvtColor(cv.imread(lenna), cv.COLOR_BGR2RGB)
subimg = img[100:400, 100:400]

#rimg, gimg, bimg = cv.split(subimg)
rimg, gimg, bimg = subimg[:, :, 0], subimg[:, :, 1], subimg[:, :, 2]

# nrow, ncol, index
plt.subplot(231), plt.imshow(img), plt.title('original')
plt.subplot(233), plt.imshow(subimg), plt.title('croped')
plt.subplot(234), plt.imshow(rimg), plt.title('red channel')
plt.subplot(235), plt.imshow(gimg), plt.title('green channel')
plt.subplot(236), plt.imshow(bimg), plt.title('blue channel')
plt.savefig(filename)

return filename

5.1.4 图片边框

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv
filename = '../static/image/2019/01/image141331.png'
lenna = '../static/image/2019/01/lenna.png'

BLUE = [255, 0, 0]
img = cv.cvtColor(cv.imread(lenna), cv.COLOR_BGR2RGB)
replicate = cv.copyMakeBorder(img, 20, 20, 20, 20, cv.BORDER_REPLICATE)
reflect = cv.copyMakeBorder(img, 20, 20, 20, 20, cv.BORDER_REFLECT)
reflect101 = cv.copyMakeBorder(img, 20, 20, 20, 20, cv.BORDER_REFLECT_101)
wrap = cv.copyMakeBorder(img, 20, 20, 20, 20, cv.BORDER_WRAP)
constant = cv.copyMakeBorder(img, 20, 20, 20, 20, cv.BORDER_CONSTANT, value=BLUE)


plt.subplot(231), plt.imshow(img, 'gray'), plt.title('ORIGINAL'), plt.xticks([]), plt.yticks([])
plt.subplot(232), plt.imshow(replicate, 'gray'), plt.title('REPLICATE'), plt.xticks([]), plt.yticks([])
plt.subplot(233), plt.imshow(reflect, 'gray'), plt.title('REFLECT'), plt.xticks([]), plt.yticks([])
plt.subplot(234), plt.imshow(reflect101, 'gray'), plt.title('REFLECT_101'), plt.xticks([]), plt.yticks([])
plt.subplot(235), plt.imshow(wrap, 'gray'), plt.title('WRAP'), plt.xticks([]), plt.yticks([])
plt.subplot(236), plt.imshow(constant, 'gray'), plt.title('CONSTANT'), plt.xticks([]), plt.yticks([])

plt.savefig(filename)
return filename

5.2 算术操作

5.2.1 图片相加

x = np.uint8([250])
y = np.uint8([10])
print('cv.add(x,y) = {}'.format(cv.add(x, y))) # 250+10 = 260 => 255
print('x + y = {}'.format(x + y))              # 250+10 = 260 % 256 => 4

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/tmp/babel-a6MpLG/python-s3kR3m", line 1, in <module>
    x = np.uint8([250])
NameError: name 'np' is not defined

5.2.2 图片混合(Blending)

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv
filename = '../static/image/2019/01/image142651.png'
lenna = cv.cvtColor(cv.imread('../static/image/2019/01/lenna.png'), cv.COLOR_BGR2RGB)
circle = np.zeros(lenna.shape).astype('uint8')
cx, cy = int(lenna.shape[0] / 2), int(lenna.shape[1] / 2)
circle = cv.circle(circle, (cx, cy), 250, (255, 255, 255), -1)

# dst = 0.7*lenna + 0.3*circle + 0
dst = cv.addWeighted(lenna, 0.7, circle, 0.3, 0)

plt.subplot(131), plt.imshow(lenna), plt.title('lenna'), plt.xticks([]), plt.yticks([])
plt.subplot(132), plt.imshow(circle), plt.title('circle'), plt.xticks([]), plt.yticks([])
plt.subplot(133), plt.imshow(dst), plt.title('dst'), plt.xticks([]), plt.yticks([])

plt.savefig(filename)
return filename

5.2.3 位操作

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv

filename = '../static/image/2019/01/image150141.png'
lenna = cv.cvtColor(cv.imread('../static/image/2019/01/lenna.png'), cv.COLOR_BGR2RGB)
roi = np.zeros(lenna.shape).astype('uint8')
cx, cy = int(lenna.shape[0] / 2), int(lenna.shape[1] / 2)
roi = cv.circle(roi, (cx, cy), 250, (255, 255, 255), -1)

andimg = cv.bitwise_and(lenna, roi)
orimg = cv.bitwise_or(lenna, roi)

plt.subplot(221), plt.imshow(lenna), plt.title('lenna'), plt.xticks([]), plt.yticks([])
plt.subplot(222), plt.imshow(roi), plt.title('roi'), plt.xticks([]), plt.yticks([])
plt.subplot(223), plt.imshow(andimg), plt.title('and'), plt.xticks([]), plt.yticks([])
plt.subplot(224), plt.imshow(orimg), plt.title('or'), plt.xticks([]), plt.yticks([])

plt.savefig(filename)
return filename

5.3 图片处理

5.3.1 几何变换

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv

filename = '../static/image/2019/01/image152231.png'
lenna = cv.cvtColor(cv.imread('../static/image/2019/01/lenna.png'), cv.COLOR_BGR2RGB)


def scale(img, scale_x, scale_y):
  h, w = img.shape[:2]
  return cv.resize(img, (int(h * scale_y), int(w * scale_x)), interpolation=cv.INTER_CUBIC)


def translation(img, trans_x, trans_y):
  h, w = img.shape[:2]
  M = np.float32([[1, 0, trans_x], [0, 1, trans_y]])
  return cv.warpAffine(img, M, (h, w))


def rotate(img, angle):
  h, w = img.shape[:2]
  center = int(w / 2.0), int(h / 2.0)
  M = cv.getRotationMatrix2D(center, angle, 1)
  return cv.warpAffine(img, M, (w, h))


def affine(img, pts1, pts2):
  h, w, d = img.shape
  M = cv.getAffineTransform(pts1, pts2)
  return cv.warpAffine(img, M, (w, h))


def perspective(img, pts1, pts2):
  h, w, d = img.shape
  M = cv.getPerspectiveTransform(pts1, pts2)
  return cv.warpPerspective(img, M, (w, h))


lenna_scale = scale(lenna, 1.5, 2)
lenna_trans = translation(lenna, 50, 100)
lenna_rotate120 = rotate(lenna, 120)

pts1 = np.float32([[50, 50], [200, 50], [50, 200]])
pts2 = np.float32([[10, 100], [200, 50], [100, 250]])
# mapping three points from origin picture to dest picture, using affine transform
lenna_affine = affine(lenna, pts1, pts2)

pts3 = np.float32([[0, 0], [512, 0], [0, 512], [512, 512]])
pts4 = np.float32([[100, 300], [350, 300], [0, 512], [512, 512]])
# mapping four points, using perspective transform
lenna_perspective = perspective(lenna, pts3, pts4)

plt.subplot(231), plt.imshow(lenna), plt.title('lenna'), plt.xticks([]), plt.yticks([])
plt.subplot(232), plt.imshow(lenna_scale), plt.title('scale'), plt.xticks([]), plt.yticks([])
plt.subplot(233), plt.imshow(lenna_trans), plt.title('translation'), plt.xticks([]), plt.yticks([])
plt.subplot(234), plt.imshow(lenna_rotate120), plt.title('rotate 120$\degree$'), plt.xticks([]), plt.yticks([])
plt.subplot(235), plt.imshow(lenna_affine), plt.title('affine'), plt.xticks([]), plt.yticks([])
plt.subplot(236), plt.imshow(lenna_perspective), plt.title('perspective'), plt.xticks([]), plt.yticks([])

plt.savefig(filename)
return filename

5.3.2 二值化(Thresholding)

普通二值化

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv
filename = '../static/image/2019/01/image190450.png'
img = cv.imread('../static/image/2019/01/lena_meyer.png', 0)

ret, thresh1 = cv.threshold(img, 127, 255, cv.THRESH_BINARY)
ret, thresh2 = cv.threshold(img, 127, 255, cv.THRESH_BINARY_INV)
ret, thresh3 = cv.threshold(img, 127, 255, cv.THRESH_TRUNC)
ret, thresh4 = cv.threshold(img, 127, 255, cv.THRESH_TOZERO)
ret, thresh5 = cv.threshold(img, 127, 255, cv.THRESH_TOZERO_INV)

titles = ['Original Image', 'BINARY', 'BINARY_INV', 'TRUNC', 'TOZERO', 'TOZERO_INV']
images = [img, thresh1, thresh2, thresh3, thresh4, thresh5]
for i in range(6):
  plt.subplot(2, 3, i+1), plt.imshow(images[i], 'gray')
  plt.title(titles[i])
  plt.xticks([]), plt.yticks([])

plt.savefig(filename)
return filename

自适应二值化

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv
filename = '../static/image/2019/01/image191045.png'
img = cv.imread('../static/image/2019/01/lena_meyer.png', 0)

img = cv.medianBlur(img, 5)
ret, th1 = cv.threshold(img, 127, 255, cv.THRESH_BINARY)
th2 = cv.adaptiveThreshold(img, 255, cv.ADAPTIVE_THRESH_MEAN_C, cv.THRESH_BINARY, 11, 2)
th3 = cv.adaptiveThreshold(img, 255, cv.ADAPTIVE_THRESH_GAUSSIAN_C, cv.THRESH_BINARY, 11, 2)

titles = [
    'Original Image', 'Global Thresholding (v = 127)',
    'Adaptive Mean Thresholding', 'Adaptive Gaussian Thresholding'
]
images = [img, th1, th2, th3]
for i in range(4):
  plt.subplot(2, 2, i+1), plt.imshow(images[i], 'gray')
  plt.title(titles[i])
  plt.xticks([]), plt.yticks([])
plt.savefig(filename)
return filename

Otsu 二值化

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv
filename = '../static/image/2019/01/image193021.png'
img = cv.imread('../static/image/2019/01/lenna.png',0)

# global thresholding
ret1,th1 = cv.threshold(img,127,255,cv.THRESH_BINARY)

# Otsu's thresholding
ret2,th2 = cv.threshold(img,0,255,cv.THRESH_BINARY+cv.THRESH_OTSU)

# Otsu's thresholding after Gaussian filtering
blur = cv.GaussianBlur(img,(5,5),0)
ret3,th3 = cv.threshold(blur,0,255,cv.THRESH_BINARY+cv.THRESH_OTSU)

images = [img, th1, th2, th3]
for i in range(4):
  plt.subplot(2,2, i+1), plt.imshow(images[i], 'gray')
  plt.xticks([]), plt.yticks([])
plt.savefig(filename)
return filename

5.3.3 平滑(Smoothing Images)

卷积

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv
filename = '../static/image/2019/01/image190747.png'

img = cv.imread('../static/image/2019/01/lena_meyer.png')
img = cv.cvtColor(img, cv.COLOR_BGR2RGB)
kernel = np.ones((5, 5), np.float32)/25
dst = cv.filter2D(img, -1, kernel)

plt.subplot(121),plt.imshow(img),plt.title('Original')
plt.xticks([]), plt.yticks([])
plt.subplot(122),plt.imshow(dst),plt.title('Averaging')
plt.xticks([]), plt.yticks([])
plt.savefig(filename)

return filename

5.3.4 Canny 边界检测

import numpy as np
import matplotlib.pyplot as plt
import cv2 as cv
filename = '../static/image/2019/01/image203420.png'
img = cv.imread('../static/image/2019/01/lena_meyer.png', 0)

edges=cv.Canny(img, 100, 200)
plt.subplot(121), plt.imshow(img, cmap='gray')
plt.title('Original Image'), plt.xticks([]), plt.yticks([])
plt.subplot(122), plt.imshow(edges, cmap='gray')
plt.title('Edge Image'), plt.xticks([]), plt.yticks([])

plt.savefig(filename)
return filename