pytorch 读取图片

以下是三种PyTorch读取图片的方法：

1. 使用`torchvision.io`库中的`image.read_image()`函数：

from torchvision.io import image

path = '../dataset/train/hazy/8902.jpg'
img = image.read_image(path)
print(type(img)) # 输出：<class 'torch.Tensor'>
print(img.shape) # 输出：torch.Size([3, 344, 550])

2. 使用`torchvision.transforms`库中的`ToTensor()`函数和`PIL`库中的`Image`类：

from torchvision.transforms import ToTensor
from PIL import Image

path = '../dataset/train/hazy/8902.jpg'
img = Image.open(path)
img = ToTensor()(img)
print(type(img)) # 输出：<class 'torch.Tensor'>
print(img.shape) # 输出：torch.Size([3, 344, 550])

3. 使用`cv2`库和`numpy`库：

import cv2
import numpy as np

path = '../dataset/train/hazy/8902.jpg'
img = cv2.imread(path)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
img = np.transpose(img, (2, 0, 1))
img = torch.from_numpy(img)
print(type(img)) # 输出：<class 'torch.Tensor'>
print(img.shape) # 输出：torch.Size([3, 344, 550])