torchvision.transforms中的PIL和Tens

torchvision.transforms模块包含了图像转换函数，这些函数可以作用于PIL对象和Tensor对象。PyTorch里面使用PIL很多，而且torchvision用PIL作为默认的后端

PIL默认为torchvision的后端

参考：《 Why is PIL used so often with Pytorch》--- PIL比OpenCV更加轻量化和效率高。《Python PIL 模块 vs OpenCV模块比较》

• PILToTensor()是把PIL Image对象转换为PyTorch Tensor对象，数据类型不变；
• ToTensor()是把取值为 [0, 255] 的PIL Image 或 numpy.ndarray 为取值为 [0.0, 1.0] 的PyTorch Tensor对象，注意：因为输入图像被缩放到 [0.0, 1.0]，所以在变换目标图像掩码时不应使用此变换。

范例：

# Import necessary libraries
import torch
from PIL import Image
import cv2
import torchvision.transforms as transforms

# Read a PIL image
image = Image.open('iceland.jpg')

# Define a transform to convert PIL
# image to a Torch tensor
transform = transforms.Compose([
    transforms.PILToTensor()
])

# transform = transforms.PILToTensor()
# Convert the PIL image to Torch tensor
img_tensor = transform(image)

# print the converted Torch tensor
print(img_tensor.shape)
print(img_tensor)

# Read the image by OpenCV
image = cv2.imread('iceland.jpg')

# Convert BGR image to RGB image
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Define a transform to convert
# the image to torch tensor
transform = transforms.Compose([
    transforms.ToTensor()
])

# Convert the image to Torch tensor
tensor = transform(image)

# print the converted image tensor
print(tensor.shape)
print(tensor)

运行结果
测试图片如下： iceland.jpg