【python计算机视觉】2、颜色空间转换

一、RGB彩色图像转灰度图像

转换公式
$$Gray=R\ast 0.3+G\ast 0.59+B\ast 0.11$$
Python代码

#输入彩色图像的四维张量，输出灰度图像的四维张量
def RGB2Gray(img):
    r,g,b = torch.split(img,1,dim=1)
    gray = torch.zeros_like(r)
    gray = 0.3*r+0.59*g+0.11*b
    return gray

二、RGB颜色空间转YCbCr颜色空间

转换公式
$$Y=0.257\ast R+0.504\ast G+0.098\ast B+16$$
$$Cb=-0.148\ast R-0.291\ast G+0.439\ast B+128$$
$$Cr=0.439\ast R-0.368\ast G-0.071\ast B+128$$
python代码

#输入RGB彩色空间的四维张量，输出YCbCr颜色空间的四维张量
def RGB2YCbCr(img):
    img = img*255.0
    r,g,b = torch.split(img,1,dim=1)
    y = torch.zeros_like(r)
    cb = torch.zeros_like(r)
    cr = torch.zeros_like(r)
    
    y = 0.257*r+0.504*g+0.098*b+16
    y = y/255.0
    
    cb = -0.148*r-0.291*g+0.439*b+128
    cb = cb/255.0
    
    cr = 0.439*r-0.368*g-0.071*b+128
    cr = cr/255.0
    
    img = torch.cat([y,cb,cr],dim=1)
    return img

三、YCbCr颜色空间转RGB颜色空间

转换公式
$$R=1.164\ast (Y-16)+1.596\ast (Cr-128)$$
$$G=1.164\ast (Y-16)-0.392\ast (Cb-128)-0.813\ast (Cr-128)$$
$$B=1.164\ast (Y-16)+2.017\ast (Cb-128)$$
python代码

#输入YCbCr颜色空间的四维张量，输出RGB颜色空间的四维张量
def YCbCr2RGB(img):
    img = img*255.0
    y,cb,cr = torch.split(img,1,dim=1)
    r = torch.zeros_like(y)
    g = torch.zeros_like(y)
    b = torch.zeros_like(y)
    
    r = 1.164*(y-16)+1.596*(cr-128)
    r = r/255.0
    g = 1.164*(y-16)-0.392*(cb-128)-0.813*(cr-128)
    g = g/255.0
    b = 1.164*(y-16)+2.017*(cb-128)
    b = b/255.0
    
    img = torch.cat([r,g,b],dim=1)
    return img