How to crop an image at random location in PyTorch

In this article, we will discuss how to pad an image on all sides in PyTorch. 

Torchvision.transforms.RandomCrop method

Cropping is a technique of removal of unwanted outer areas from an image to achieve this we use a method in python that is torchvision.transforms.RandomCrop(). It is used to crop an image at a random location in PyTorch. This method accepts images like PIL Image and Tensor Image. The tensor image is a PyTorch tensor with [C, H, W] shape, where C represents a number of channels and  H, W represents height and width respectively.

Syntax:  torchvision.transforms.RandomCrop(size)

Parameters:

• size: Desired crop size of the image.

Return: it returns the cropped image of given input size.

Image used for demonstration:

Example 1:

In this example, we are transforming the image with a height of 200 and a width of 400.

Python3

# import required libraries
import torch
import torchvision.transforms as transforms
from PIL import Image

# Read image
image = Image.open('pic.jpg')

# create an transform for crop the image
# 200px height and 400px wide
transform = transforms.RandomCrop((200, 400))

# use above created transform to crop
# the image
image_crop = transform(image)

# display result
image_crop.show()

Output:

Example 2:

In this example, we are transforming the image at the center. In this, we will get a square image as output.

Python3

# import required libraries
import torch
import torchvision.transforms as transforms
from PIL import Image

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

# create an transform for crop the image
transform = transforms.RandomCrop(300)

# use above created transform to crop
# the image
image_crop = transform(image)

# display result
image_crop.show()

Output: