Python is now one of the most widely used programming languages in the world. It can be used to perform various functions and tasks using a simple syntax.
One important function Python can perform is automatic processing of digital images, which can be done using Pillow.
In this tutorial, we’ll show you how to process images using the Pillow module. Then, we’ll take it a step further and demonstrate how to implement some basic image operations.
To follow along with this tutorial, you should have basic knowledge of Python and the Python 3 interpreter installed on your local machine.
What is Pillow?
Pillow is a fork of the Python Imaging Library (PIL). It is a free and open-source library for manipulating and processing images.
PIL is a powerful library in its own right, but it hasn’t been updated since 2009 and doesn’t support Python 3. Pillow provides more features and support for Python 3.
Pillow supports a range of image file formats, such as .PNG, .JPEG, .PPM, .GIF, .TIFF, and .BMP. You can perform various operations on images such as cropping, resizing, adding text, rotating, grayscaling, and so much more using this library.
Installation and project setup
You can install Pillow using pip, a package manager for Python packages:
python3 -m pip install --upgrade pip python3 -m pip install --upgrade Pillow
Pillow offers the
Image object, which has inbuilt functions and properties on which manipulation operations can be carried out.
To get started, first import the
I``mage object to the Python file.
from PIL import Image
Next, load the image by calling the
Image.open() function, which returns a value of the
Image object data type.
image = Image.open('sample.jpg')
For our examples, we’ll use a sample image from Unsplash.
Also worth noting, the images are in the same directory as the Python script file being run.
Properties of the
There are several properties of the image we can access to get more data from the image:
image.widthreturns the width of the image
image.heightreturns the height of the image
image.formatreturns the file format of the image (e.g., .JPEG, .BMP, .PNG, etc.)
image.sizereturns the tuple height and weight of the image
image.palettereturns the color palette table, if one exists
image.modereturns the pixel format of the image (e.g., 1, L, RGB, CMYK)
Basic image operations
We are also able to process and manipulate our images using various operations.
Any changes made to the
Image object can be saved to an image file with the
save() method. All the rotations, resizing, cropping, drawing, and other image manipulations are done through via calls on this
Let’s zoom in and explore some of these operations in more detail.
Changing image formats
Pillow supports a wide variety of images formats. An image can be converted from one format to another as follows:
image = Image.open('sample.jpg') image.save('sample_formatted.png')
First, the image is loaded. Then, Pillow sees the file extension specified as
PNG, so it converts the image to .PNG before saving it to file.
You can resize images by creating a thumbnail of the image using Pillow.
thumbnail() function, the image is resized to keep its aspect ratio. This takes two values representing the maximum width and maximum height of the thumbnail.
image = Image.open('sample.jpg') image.thumbnail((200, 200)) image.save('sample_thumbnail.jpg')
The image is then resized within the maximum limit so as not to be overstretched or blurred.
Flipping and rotating images
If you need the image to face a different direction, Pillow enables you to flip it. This is done using the
transpose function, which takes any of the following parameters:
Image.FLIP_LEFT_RIGHT, which flips the image horizontally
Image.FLIP_TOP_BOTTOM, which flips the image vertically
Image.ROTATE_90, which rotates the image to a certain degree, depending on the angle
image = Image.open('sample.jpg') image.transpose(Image.FLIP_TOP_BOTTOM) image.save('sample_flip.jpg')
The resulting image is flipped vertically.
Alternatively, you can rotate images using the
rotate() method. This takes an integer or float argument representing the degrees of rotation and returns a new
Image object of the rotated image. The rotation is counterclockwise.
image = Image.open('sample.jpg') image.rotate(90) image.save('image_rotate90.jpg')
The image is rotated by an angle of 90 degrees.
To crop an image is to cut out only a specific portion. Cropping often comes into play when editing images for web applications.
crop() function in Pillow requires the portion to be cropped as a rectangle. The method takes a box tuple that defines the position and size of the cropped region and returns an
Image object representing the cropped image. The region is defined by a 4-tuple, where coordinates are (left, upper, right, lower).
image = Image.open('sample.jpg') image.crop(200, 50, 450, 300) image.save('sample_cropped.jpg')
In the example above, the first two values represent the starting position from the upper-left; the third and fourth values represent the distance in pixels from the starting position toward the right and bottom direction.
The full size of the cropped image can be calculated as 250×250 pixels.
There are various forms of pixel representations, including L (luminance), RGB, and CMYK.
Pillow allows you to convert images between different pixel representations using the
convert() method. The library supports transformations between each supported mode as well as the “L” and “RGB” modes. To convert between other modes, you may have to use an “RGB” image.
image = Image.open('sample.jpg') grayscale_image = image.convert('L') grayscale_image.save('sample_grayscale.jpg')
Using the convert function, the sample image is converted from RGB to L (luminance) mode, which will result in a grayscale image.
The act of modifying and enhancing images to improve the appearance can be known as filtering.
ImageFilter module from Pillow, you can access to the filter() method and use various filter techniques, including:
For example, let’s take a look at the
from PIL import Image, ImageFilter image = Image.open('sample.jpg') edges_image = image.filter(ImageFilter.FIND_EDGES) edges_image.save('sample_edges.jpg')
The filter processes the image to display the edges of the image.
Any image can be processed using any of the filters available to produce the desired output.
Processing images with Pillow: A practical example
Now that we have a basic understanding of the library, let’s create a simple python script to automate the processing of various types of images.
Let’s say you’re given a group of images and asked to add a watermark to each image.
To solve the problem, you can create a Python file called
script.py in the same folder as the images.
First, import all necessary modules:
import os from PIL import Image
The OS module in Python provides functions for creating and removing a directory and changing and identifying the current directory.
Create a directory for the processed image:
Store the width and height of the logo image:
logo_image = Image.open('watermark_logo.png') logo_image = logo_image.resize((50, 50)) logo_width, logo_height = logo_image.size
os.listdir function together with a for loop:
for image in os.listdir('./images'): try: # Separting the filepath from the image's name path, filename = os.path.split(image) filename = os.path.splitext(filename)
Open the image:
image = Image.open('./images/'+image) #Resizing the image to a set size. edited_image = image.resize((300, 300)) #Setting the position for the placement width = edited_image.width height = edited_image.height
Use the paste function to position the logo on the image:
edited_image.paste(logo_image, (width - logo_width, height - logo_height), logo_image)
Save the images in the new directory:
edited_image.save('./watermarked_Images/' + filename + ".jpg")
Each image in the directory has been processed and the watermark added. This script enabled us to efficiently perform the task in less time.
You can also check out the GitHub repository for the full code and resources associated with this tutorial.
Pillow is a powerful library for processing images in Python. In this tutorial, we established a foundational understanding of how to perform basic operations on images in Python using Pillow.
If you want to learn more, check out the official Pillow documentation.
Cut through the noise of traditional error reporting with LogRocket
LogRocket is a digital experience analytics solution that shields you from the hundreds of false-positive errors alerts to just a few truly important items. LogRocket tells you the most impactful bugs and UX issues actually impacting users in your applications.
Then, use session replay with deep technical telemetry to see exactly what the user saw and what caused the problem, as if you were looking over their shoulder.
LogRocket automatically aggregates client side errors, JS exceptions, frontend performance metrics, and user interactions. Then LogRocket uses machine learning to tell you which problems are affecting the most users and provides the context you need to fix it.
Focus on the bugs that matter — try LogRocket today.