Python+OpenCV实现图像融合的原理及代码_Python

Python+OpenCV实现图像融合的原理及代码

2021-04-24 00:49～沐春风～ Python

这篇文章主要介绍了Python+OpenCV实现图像融合的原理及代码,本文通过实例代码给大家介绍的非常详细，具有一定的参考借鉴价值，需要的朋友可以参考下

根据导师作业安排，在学习数字图像处理（刚萨雷斯版）第六章彩色图像处理中的彩色模型后，导师安排了一个比较有趣的作业：

Python+OpenCV实现图像融合的原理及代码

融合原理为：

1 注意：遥感原rgb图image和灰度图grayimage为测试用的输入图像；

2 步骤：（1）将rgb转换为hsv空间（h：色调，s：饱和度，v：明度）；

（2）用gray图像诶换掉hsv中的v；

（3）替换后的hsv转换回rgb空间即可得到结果。

书上只介绍了hsi彩色模型，并没有说到hsv，所以需要网上查找资料。

python代码如下：

100

101

102

103

104

105

106

107

108

109

110

									import cv2

									import numpy as np

									import math

									from matplotlib import pyplot as plt

									def caijian(img):#裁剪图像与否根据选择图像大小而定，调用了opencv函数

									weight=img.shape[0]

									height=img.shape[1]

									print(“图像大小为：%d*%d”%(weight,height))

									img=cv2.resize(img,(int(weight/2),int(height/2)),interpolation=cv2.inter_cubic)

									return(img)

									def graytograyimg(img):

									grayimg=img1

									weight=img.shape[0]

									height=img.shape[1]

									for i in range(weight):

									for j in range(height):

									grayimg[i,j]=0.299img[i,j,0]+0.587img[i,j,1]+0.114img[i,j,2]

									return(grayimg)

									def rgbtohsv(img):

									b,g,r=cv2.split(img)

									rows,cols=b.shape

									h=np.ones([rows,cols],“float”)

									s=np.ones([rows,cols],“float”)

									v=np.ones([rows,cols],“float”)

									print(“rgb图像大小：%d*%d”%(rows,cols))

									for i in range(0, rows):

									for j in range(0, cols):

									max=max((b[i,j],g[i,j],r[i,j]))

									min=min((b[i,j],g[i,j],r[i,j]))

									v[i,j]=max

									if v[i,j]0:

									s[i,j]=0

									else:

									s[i,j]=(v[i,j]-min)/v[i,j]

									if maxmin:

									h[i,j]=0 # 如果rgb三向量相同，色调为黑

									elif v[i,j]==r[i,j]:

									h[i,j]=(60*(float(g[i,j])-b[i,j])/(v[i,j]-min))

									elif v[i,j]==g[i,j]:

									h[i,j]=60*(float(b[i,j])-r[i,j])/(v[i,j]-min)+120

									elif v[i,j]==b[i,j]:

									h[i,j]=60*(float(r[i,j])-g[i,j])/(v[i,j]-min)+240

									if h[i,j]<0:

									h[i,j]=h[i,j]+360

									h[i,j]=h[i,j]/2

									s[i,j]=255*s[i,j]

									result=cv2.merge((h,s,v)) # cv2.merge函数是合并单通道成多通道

									result=np.uint8(result)

									return(result)

									def graytohsgry(grayimg,hsvimg):

									h,s,v=cv2.split(hsvimg)

									rows,cols=v.shape

									for i in range(rows):

									for j in range(cols):

									v[i,j]=grayimg[i][j][0]

									newimg=cv2.merge([h,s,v])

									newimg=np.uint8(newimg)

									return newimg

									def hsvtorgb(img,rgb):

									h1,s1,v1=cv2.split(img)

									rg = rgb.copy()

									rows,cols=h1.shape

									r,g,b=0.0,0.0,0.0

									b1,g1,r1 = cv2.split(rg)

									print(“hsv图像大小为：%d*%d”%(rows,cols))

									for i in range(rows):

									for j in range(cols):

									h=h1[i][j]

									v=v1[i][j]/255

									s=s1[i][j]/255

									h=h2

									hx=int(h/60.0)

									hi=hx%6

									f=hx-hi

									p=v(1-s)

									q=v*(1-fs)

									t=v(1-(1-f)s)

									if hi0:

									r,g,b=v,t,p

									elif hi1:

									r,g,b=q,v,p

									elif hi2:

									r,g,b=p,v,t

									elif hi3:

									r,g,b=p,q,v

									elif hi4:

									r,g,b=t,p,v

									elif hi5:

									r,g,b=v,p,q

									r,g,b=(r255),(g255),(b255)

									r1[i][j]=int®

									g1[i][j]=int(g)

									b1[i][j]=int(b)

									rg=cv2.merge([b1,g1,r1])

									return rg

									img=cv2.imread(“d:/rgb.bmp”)

									gray=cv2.imread(“d:/gray.bmp”)

									img=caijian(img)

									gray=caijian(gray)

									grayimg=graytograyimg(gray)

									hsvimg=rgbtohsv(img)

									hsgray=graytohsgry(grayimg,hsvimg)

									rgbimg=hsvtorgb(hsgray,img)

									cv2.imshow(“image”,img)

									cv2.imshow(“grayimage”,grayimg)

									cv2.imshow(“hsvimage”,hsvimg)

									cv2.imshow(“hsgrayimage”,hsgray)

									cv2.imshow(“rgbimage”,rgbimg)

									cv2.waitkey(0)

									cv2.destroyallwindows()