脚本之家,脚本语言编程技术及教程分享平台!
分类导航

Python|VBS|Ruby|Lua|perl|VBA|Golang|PowerShell|Erlang|autoit|Dos|bat|

服务器之家 - 脚本之家 - Python - Python-numpy实现灰度图像的分块和合并方式

Python-numpy实现灰度图像的分块和合并方式

2020-05-01 11:17weixin_38671006 Python

今天小编就为大家分享一篇Python-numpy实现灰度图像的分块和合并方式,具有很好的参考价值,希望对大家有所帮助。一起跟随小编过来看看吧

我就废话不多说了,直接上代码吧!

?
1
2
3
4
5
6
7
8
9
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
from numpy import *
import numpy as np
import cv2, os, math, os.path
from PIL import Image
base="F:\\Spy_CNN\\pythonCode\\cvSPY\\cvTest\\LBP\\LBPImag3\\"
base2="F:\\ProgrameCode\\FaceDataLib\\orl_Arry\\"
imageOld=cv2.imread(base2+"s1_1.bmp")
image=cv2.cvtColor(imageOld,cv2.COLOR_BGR2GRAY)
'''图像的合成'''
H,W=image.shape#(112, 92)
kuai=5
a=1#为了好调程序
maskx,masky = H/kuai,W/kuai  #29 14
toImage=np.zeros((H+(kuai-1)*a,W+(kuai-1)*a))
toImage.shape
 
#您画画图,总结规律,就可以想出来了
 
for i in range(kuai):
    for j in range(kuai):
        '''float64 array'''
        faceZi=image[int(i*maskx): int((i+1)*maskx),int(j*masky) :int((j+1)*masky)]
        cv2.imwrite(base+str(i)+str(j)+".bmp",faceZi)
#        toImage[int(i*maskx)+a: int((i+1)*maskx)+a,int(j*masky)+a :int((j+1)*masky)+a]=faceZi  
        toImage[int(i*maskx)+i: int((i+1)*maskx)+i,int(j*masky)+j :int((j+1)*masky)+j]=faceZi            
cv2.imwrite(base+"toImage.bmp",toImage)
#最简单的,直接在灰度图像上画出一条黑线,只是会丢失一些像素的数据
for i in range(1,kuai):
    print(i)
    toImage[int(i*maskx),:]=0
    toImage[:,int(i*masky)]=0
cv2.imwrite(base+"toImage.bmp",toImage)
 
 

以上这篇Python-numpy实现灰度图像的分块和合并方式就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持服务器之家。

原文链接:https://blog.csdn.net/weixin_38671006/article/details/92787612

延伸 · 阅读

精彩推荐