深度学习中对图片和标签执行相同位置裁剪及其他变换操作

1.代码1

import torchvision.transforms.functional as TF
import random
from torchvision.transforms import transforms as tfs
def rand_crop(data,label,height,width):
    h,w,_=data.shape
    top=random.randint(0,h-height)
    left=random.randint(0,w-width)
    data=data[top:top+height,left:left+width]
    label=label[top:top+height,left:left+width]
    return data,label

2.代码2

import torchvision.transforms.functional as TF

class MyDataset(Dataset):
    def __init__(self, image_paths, target_paths, train=True):
        self.image_paths = image_paths
        self.target_paths = target_paths

    def transform(self, image, mask):
        # Resize
        resize = transforms.Resize(size=(520, 520))
        image = resize(image)
        mask = resize(mask)

        # Random crop
        i, j, h, w = transforms.RandomCrop.get_params(
            image, output_size=(512, 512))
        image = TF.crop(image, i, j, h, w)
        mask = TF.crop(mask, i, j, h, w)

        # Random horizontal flipping
        if random.random() > 0.5:
            image = TF.hflip(image)
            mask = TF.hflip(mask)

        # Random vertical flipping
        if random.random() > 0.5:
            image = TF.vflip(image)
            mask = TF.vflip(mask)

        # Transform to tensor
        image = TF.to_tensor(image)
        mask = TF.to_tensor(mask)
        return image, mask

    def __getitem__(self, index):
        image = Image.open(self.image_paths[index])
        mask = Image.open(self.target_paths[index])
        x, y = self.transform(image, mask)
        return x, y

    def __len__(self):
        return len(self.image_paths)

3.代码3

from torchvision.transforms import transforms as tfs
def rand_crop(data, label,  height, width):
  crop_size=[height,width]
  i, j, h, w = tfs.RandomCrop.get_params(data, output_size=crop_size)
  data = TF.crop(data, i, j, h, w)
  label = TF.crop(label, i, j, h, w)
  return data, label

4.代码4

from mxnet import gluon, image
def rand_crop(data, label, height, width):
 '''
# data is PIL.Image object
# label is PIL.Image object
# '''
 data=mxnet.nd.array(data)
 label=mxnet.nd.array(label)
 data, rect =image.random_crop(data,(width, height))
 label = image.fixed_crop(label,*rect)
 data=data.asnumpy()
 label=label.asnumpy()
#return data, label