Real Python: Image Processing With the Python Pillow Library

When you look at an image, you see the objects and people in it. However, when you read an image programmatically with Python or any other language, the computer sees an array of numbers. In this tutorial, you’ll learn how to manipulate images and perform basic image processing using the Python Pillow library.

Pillow and its predecessor, PIL, are the original Python libraries for dealing with images. Even though there are other Python libraries for image processing, Pillow remains an important tool for understanding and dealing with images.

To manipulate and process images, Pillow provides tools that are similar to ones found in image processing software such as Photoshop. Some of the more modern Python image processing libraries are built on top of Pillow and often provide more advanced functionality.

In this tutorial, you’ll learn how to:

• Read images with Pillow
• Perform basic image manipulation operations
• Use Pillow for image processing
• Use NumPy with Pillow for further processing
• Create animations using Pillow

This tutorial provides an overview of what you can achieve with the Python Pillow library through some of its most common methods. Once you gain confidence using these methods, then you can use Pillow’s documentation to explore the rest of the methods in the library. If you’ve never worked with images in Python before, this is a great opportunity to jump right in!

In this tutorial, you’ll use several images, which you can download from the tutorial’s image repository:

With those images in hand, you’re now ready to get started with Pillow.

Basic Image Operations With the Python Pillow Library

The Python Pillow library is a fork of an older library called PIL. PIL stands for Python Imaging Library, and it’s the original library that enabled Python to deal with images. PIL was discontinued in 2011 and only supports Python 2. To use its developers’ own description, Pillow is the friendly PIL fork that kept the library alive and includes support for Python 3.

There’s more than one module in Python to deal with images and perform image processing. If you want to deal with images directly by manipulating their pixels, then you can use NumPy and SciPy. Other popular libraries for image processing are OpenCV, scikit-image, and Mahotas. Some of these libraries are faster and more powerful than Pillow.

However, Pillow remains an important tool for dealing with images. It provides image processing features that are similar to ones found in image processing software such as Photoshop. Pillow is often the preferred option for high-level image processing tasks that don’t require more advanced image processing expertise. It’s also often used for exploratory work when dealing with images.

Pillow also has the advantage of being widely used by the Python community, and it doesn’t have the same steep learning curve as some of the other image processing libraries.

You’ll need to install the library before you can use it. You can install Pillow using pip within a virtual environment:

• Windows
• Linux + macOS

PS> python-mvenvvenvPS> .\venv\Scripts\activate(venv)PS> python-mpipinstallPillow

$ python -m venv venv
$ source venv/bin/activate
(venv)$ python -m pip install Pillow

Now that you’ve installed the package, you’re ready to start familiarizing yourself with the Python Pillow library and perform basic manipulations of images.

The Image Module and Image Class in Pillow

The main class defined in Pillow is the Image class. When you read an image using Pillow, the image is stored in an object of type Image.

For the code in this section, you’ll need the image file named buildings.jpg (image credit), which you can find in the image repository for this tutorial:

You can place this image file in the project folder that you’re working in.

When exploring images with Pillow, it’s best to use an interactive REPL environment. You’ll start by opening the image that you just downloaded:

>>> fromPILimportImage>>> filename="buildings.jpg">>> withImage.open(filename)asimg:... img.load()...>>> type(img)<class 'PIL.JpegImagePlugin.JpegImageFile'>>>> isinstance(img,Image.Image)True