OpenCV:数组矩阵

ndarray对象

def array(p_object, dtype=None, *args, **kwargs):

a=np.array([1,2,3],ndmin=2)

a=np.array([1,2,3],dtype=complex)

NumPy数据类型

numpy 的数值类型实际上是 dtype 对象的实例，并对应唯一的字符，包括 np.bool_，np.int32，np.float32，等等。

新建数组

x=np.empty([3],dtype=int)
print(x)

为随机值，未初始化

x=np.zeros([3,2],dtype=int)
print(x)

x=np.ones([3,2],dtype=int)
print(x)

从已有的数组创建数组

import numpy as np
x=[[1,2,3],[2,3,2]]
a=np.asarray(x)
print(a.shape)

import numpy as np
str=b'hello'
a=np.frombuffer(str,dtype='S1')
print(a)

import numpy as np
list=range(20)
it=iter(list)
x=np.fromiter(it,dtype=int)
print(x)

从数值范围创建数组

x=np.arange(12,dtype=float)
print(x)

import numpy as np
x=np.linspace(0,9,10,dtype=int)
print(x)

import numpy as np
x=np.logspace(0,9,10,base=2)
print(x)

切片与索引

import numpy as np
x=np.arange(15)
a=slice(2,10,3)
print(x[a])

import numpy as np
x=np.arange(15)
a=x[2:12:3]
print(a)

import numpy as np
x=np.array([[1,2,3],[4,5,6],[7,8,9]])
print(x[...,1])#第二列
print(x[1,...])#第二行
print(x[...,1:])#第二列及以下

高级索引

整数数组索引

import numpy as np
#获取(0,0),(1,1),(2,0)
x=np.array([[1,2],[3,4],[5,6]])
y=x[[0,1,2],[0,1,0]]
print(y)      
/*
[1 4 5]
*/

布尔索引

import numpy as np
x=np.arange(1,50)
print(x[x>10])    
/*
[11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49]
*/

花式索引

import numpy as np
x=np.arange(32).reshape(4,8)
print(x[[0,2,1]])    
/*
[[ 0  1  2  3  4  5  6  7]
 [16 17 18 19 20 21 22 23]
 [ 8  9 10 11 12 13 14 15]]
*/ 

迭代数组

迭代对象nditer

import numpy as np
x=np.arange(6).reshape(3,2)
for i in np.nditer(x):
    print(i,end=",")

控制遍历顺序

import numpy as np
x=np.arange(6).reshape(3,2)
for i in np.nditer(x,order='F'):
    print(i,end=",")     
/*
0,2,4,1,3,5,
*/

import numpy as np
x=np.arange(6).reshape(3,2)
for i in np.nditer(x,order='C'):
    print(i,end=",")      
/*
0,1,2,3,4,5,
*/

修改数组中元素的值

import numpy as np
x=np.arange(0,60,5).reshape(3,4)
print(x)
for i in np.nditer(x,op_flags=['readwrite']):
    i[...]=2*i
print(x)

使用外部循环

import numpy as np
x=np.arange(0,60,5).reshape(3,4)
for i in np.nditer(x,flags=['external_loop'],order='F'):
    print(i,end=',')    
/*
[ 0 20 40],[ 5 25 45],[10 30 50],[15 35 55],
*/

广播迭代

import numpy as np
x=np.arange(0,60,5).reshape(3,4)
b=np.array([1,2,3,4],dtype=int)
for i,j in np.nditer([x,b]):
    print(i,j)

数组操作

修改数组形状

import numpy as np
x=np.arange(0,60,5).reshape(3,4)
print(x)
for j in x.flat:
    print(j,end=" ")

反转数组

import numpy as np
x=np.arange(0,60,5).reshape(3,4)
print(np.transpose(x))

这篇好文章是转载于：学新通技术网