计算机视觉基础实践
环境安装
手动安装第三方库。
打开anaconda prompt
pip install opencv_python-3.4.1.15-cp36-cp36m-win_amd64.whl
pip install opencv_contrib_python-3.4.1.15-cp36-cp36m-win_amd64.whl
cd Scripts
pip install opencv_contrib_python-3.4.1.15-cp36-cp36m-win_amd64.whl
检查cv2版本
python
import cv2
cv2.__verison__
'3.4.1'
注意:cp36-cp36m对应python版本为3.6
图像基本操作
- 读取图像:cv2.imread(’ ')
- cv2.IMREAD_COLOR:彩色图像
- cv2.IMREAD_GRAYSCALE:灰度图像
- 展示图象:cv2.imshow()
- 属性:
- img.shape: 长宽通道-BGR
- 颜色通道提取:
b,g,r=cv2.split(img)
- 保存:cv2.imwrite()
- 图片截取
img=cv2.imread('cat.jpg')
cat=img[0:50,0:200]
cv_show('cat',cat)
- 边界填充
top_size,bottom_size,left_size,right_size = (50,50,50,50)
replicate = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size, borderType=cv2.BORDER_REPLICATE)
reflect = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size,cv2.BORDER_REFLECT)
reflect101 = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size, cv2.BORDER_REFLECT_101)
wrap = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size, cv2.BORDER_WRAP)
constant = cv2.copyMakeBorder(img, top_size, bottom_size, left_size, right_size,cv2.BORDER_CONSTANT, value=0)
函数语法:copyMakeBorder (src, dst, top, bottom, left, right, borderType, value)
参数解释
Src
输入矩阵
Dst
输出矩阵,对src边界扩充后的结果
Top
上侧扩充的行数
Bottom
下侧扩充的行数
Left
左侧扩充的行数
Right
右侧扩充的行数
borderType
(边界扩充类型)
BORDER_CONSTANT:常数扩充
BORDER_REPLICATE:边界复制
BORDER_REFLECT:反射扩充
BORDER_WRAP:平铺扩充
BORDER_REFLECT_101:以边界为中心反射扩充
BORDER_REFLECT101=BORDER_REFLECT_101
BORDER_DEFAULT=BORDER_REFLECT_101(这个经常是默认)
Value
borderType=BORDER_CONSTANT时填充的常数(或向量)
读取视频
# 检查是否打开正确
if vc.isOpened():
# open代表着这一帧是否被读取成功
# frame是帧
oepn, frame = vc.read()
else:
open = False
while open:
ret, frame = vc.read()
if frame is None:
break
if ret == True:
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
cv2.imshow('result', gray)
if cv2.waitKey(100) & 0xFF == 27:# 规定退出按键
break
vc.release()
cv2.destroyAllWindows()
阈值处理
ret, dst = cv2.threshold(src, thresh, maxval, type)
-
src: 输入图,只能输入单通道图像,通常来说为灰度图
-
dst: 输出图
-
thresh: 阈值
-
maxval: 当像素值超过了阈值(或者小于阈值,根据type来决定),所赋予的值
-
type:二值化操作的类型,包含以下5种类型: cv2.THRESH_BINARY; cv2.THRESH_BINARY_INV; cv2.THRESH_TRUNC; cv2.THRESH_TOZERO;cv2.THRESH_TOZERO_INV
-
cv2.THRESH_BINARY 超过阈值部分取maxval(最大值),否则取0
-
cv2.THRESH_BINARY_INV THRESH_BINARY的反转
-
cv2.THRESH_TRUNC 大于阈值部分设为阈值,否则不变
-
cv2.THRESH_TOZERO 大于阈值部分不改变,否则设为0
-
cv2.THRESH_TOZERO_INV THRESH_TOZERO的反转
ret, thresh1 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_BINARY)
ret, thresh2 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_BINARY_INV)
ret, thresh3 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_TRUNC)
ret, thresh4 = cv2.threshold(img_gray, 127, 255, cv2.THRESH_TOZERO)
ret, thresh5 = cv2.threshold(img_gray, 127, 255, cv2.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.show()
图像平滑
- 均值滤波
img = cv2.imread('lenaNoise.png')
cv2.imshow('img', img)
cv2.waitKey(0)
cv2.destroyAllWindows()
- 方框滤波
# 基本和均值一样,可以选择归一化
box = cv2.boxFilter(img,-1,(3,3), normalize=True)
cv2.imshow('box', box)
cv2.waitKey(0)
cv2.destroyAllWindows()
# 基本和均值一样,可以选择归一化,容易越界
box = cv2.boxFilter(img,-1,(3,3), normalize=False)
cv2.imshow('box', box)
cv2.waitKey(0)
cv2.destroyAllWindows()
- 高斯滤波
aussian = cv2.GaussianBlur(img, (5, 5), 1)
cv2.imshow('aussian', aussian)
cv2.waitKey(0)
cv2.destroyAllWindows()
- 中值滤波
median = cv2.medianBlur(img, 5) # 中值滤波
cv2.imshow('median', median)
cv2.waitKey(0)
cv2.destroyAllWindows()
- 展示所有滤波结果
# 展示所有的
res = np.hstack((blur,aussian,median))
# hstack是将图像水平排列;vhstack是竖直排列
#print (res)
cv2.imshow('median vs average', res)
cv2.waitKey(0)
cv2.destroyAllWindows()
形态学腐蚀与膨胀
腐蚀和膨胀是对白色部分(高亮部分)而言的,不是黑色部分。膨胀就是图像中的高亮部分进行膨胀,“领域扩张”,效果图拥有比原图更大的高亮区域。腐蚀就是原图中的高亮部分被腐蚀,“领域被蚕食”,效果图拥有比原图更小的高亮区域。
腐蚀
kernel = np.ones((3,3),np.uint8)
dige_erosion = cv2.erode(img,kernel,iterations = 1)
cv2.imshow('erosion', erosion)
cv2.waitKey(0)
cv2.destroyAllWindows()
膨胀
kernel = np.ones((3,3),np.uint8)
dige_dilate = cv2.dilate(dige_erosion,kernel,iterations = 1)
cv2.imshow('dilate', dige_dilate)
cv2.waitKey(0)
cv2.destroyAllWindows()
开运算与闭运算
开:先腐蚀,再膨胀
去掉一些腐蚀带来的负面影响
img = cv2.imread('dige.png')
kernel = np.ones((5,5),np.uint8)
opening = cv2.morphologyEx(img, cv2.MORPH_OPEN, kernel)
cv2.imshow('opening', opening)
cv2.waitKey(0)
cv2.destroyAllWindows()
闭:先膨胀,再腐蚀
边缘效应还会存在
img = cv2.imread('dige.png')
kernel = np.ones((5,5),np.uint8)
closing = cv2.morphologyEx(img, cv2.MORPH_CLOSE, kernel)
cv2.imshow('closing', closing)
cv2.waitKey(0)
cv2.destroyAllWindows()
梯度运算
梯度=膨胀-腐蚀
将边缘检测出来
pie = cv2.imread('pie.png')
kernel = np.ones((7,7),np.uint8)
dilate = cv2.dilate(pie,kernel,iterations = 5)
erosion = cv2.erode(pie,kernel,iterations = 5)
res = np.hstack((dilate,erosion))
cv2.imshow('res', res)
cv2.waitKey(0)
cv2.destroyAllWindows()
gradient = cv2.morphologyEx(pie, cv2.MORPH_GRADIENT, kernel)
cv2.imshow('gradient', gradient)
cv2.waitKey(0)
cv2.destroyAllWindows()
梯度
Sobel 算子
G
x
=
[
−
1
0
1
−
2
0
2
−
1
0
1
]
∗
A
G_x=\left[ \begin{matrix}-1 &0&1\\-2&0&2\\-1&0&1 \end{matrix} \right]*A
Gx=⎣⎡−1−2−1000121⎦⎤∗A
G
y
=
[
−
1
−
2
−
1
0
0
0
1
2
1
]
∗
A
G_y=\left[ \begin{matrix}-1 &-2&-1\\0&0&0\\1&2&1 \end{matrix} \right]*A
Gy=⎣⎡−101−202−101⎦⎤∗A
dst = cv2.Sobel(src, ddepth, dx, dy, ksize)
- ddepth:图像的深度
- dx和dy分别表示水平和竖直方向
- ksize是Sobel算子的大小
白到黑是正数,黑到白就是负数了,所有的负数会被截断成0,所以要取绝对值
# CV_64F作用是保持负数输出
sobelx = cv2.Sobel(img,cv2.CV_64F,1,0,ksize=3)
# 1代表着x方向
# 将负数转化成绝对值
sobelx = cv2.convertScaleAbs(sobelx)
分别计算x和y,再求和,不建议直接计算
sobely = cv2.Sobel(img,cv2.CV_64F,0,1,ksize=3)
# 1代表着y方向
sobely = cv2.convertScaleAbs(sobely)
# 按权重相加
sobelxy = cv2.addWeighted(sobelx,0.5,sobely,0.5,0)
img = cv2.imread('lena.jpg',cv2.IMREAD_GRAYSCALE)
sobelx = cv2.Sobel(img,cv2.CV_64F,1,0,ksize=3)
sobelx = cv2.convertScaleAbs(sobelx)
sobely = cv2.Sobel(img,cv2.CV_64F,0,1,ksize=3)
sobely = cv2.convertScaleAbs(sobely)
sobelxy = cv2.addWeighted(sobelx,0.5,sobely,0.5,0)
图像梯度-Scharr算子
G
x
=
[
−
3
0
3
−
10
0
10
−
3
0
3
]
∗
A
G_x=\left[ \begin{matrix}-3&0&3\\-10&0&10\\-3&0&3 \end{matrix} \right]*A
Gx=⎣⎡−3−10−30003103⎦⎤∗A
G
y
=
[
−
3
−
10
−
3
0
0
0
3
−
10
−
3
]
∗
A
G_y=\left[ \begin{matrix}-3&-10&-3\\0&0&0\\3&-10&-3 \end{matrix} \right]*A
Gy=⎣⎡−303−100−10−30−3⎦⎤∗A
图像梯度-laplacian算子
G
x
=
[
0
1
0
1
4
1
0
1
0
]
G_x=\left[ \begin{matrix}0&1&0\\1&4&1\\0&1&0 \end{matrix} \right]
Gx=⎣⎡010141010⎦⎤
对噪声敏感
Canny 边缘检测
-
1使用高斯滤波器,以平滑图像,滤除噪声。
-
2计算图像中每个像素点的梯度强度和方向。
-
3应用非极大值(Non-Maximum Suppression)抑制,以消除边缘检测带来的杂散响应。
-
4应用双阈值(Double-Threshold)检测来确定真实的和潜在的边缘。
-
5通过抑制孤立的弱边缘最终完成边缘检测。
1.高斯滤波器
2.梯度和方向
3.非极大值抑制
4.双阈值检测
代码实现canny
img=cv2.imread("lena.jpg",cv2.IMREAD_GRAYSCALE)
v1=cv2.Canny(img,80,150)
v2=cv2.Canny(img,50,100)