The Keras deep learning library provides a sophisticated API for loading, preparing, and augmenting image data.
Also included in the API are some undocumented functions that allow you to quickly and easily load, convert, and save image files.
These functions can be convenient when getting started on a computer vision deep learning project, allowing you to use the same Keras API initially to inspect and handle image data.
In this tutorial, you will discover how to use the basic image handling functions provided by the Keras API.
After completing this tutorial, you will know:Let’s get started.
This tutorial is divided into five parts; they are:The first step is to select a test image to use in this tutorial.
We will use a photograph of Bondi Beach, Sydney, taken by Isabell Schulz, released under a permissive creative commons license.
Bondi Beach, SydneyDownload the image and place it into your current working directory with the filename “bondi_beach.
jpg“.
The Keras deep learning library provides utilities for working with image data.
The main API is the ImageDataGenerator class that combines data loading, preparation, and augmentation.
We will not cover the ImageDataGenerator class in this tutorial.
Instead, we will take a closer look at a few less-documented or undocumented functions that may be useful when working with image data and modeling with the Keras API.
Specifically, Keras provides functions for loading, converting, and saving image data.
The functions are in the utils.
py function and exposed via the image.
py module.
These functions can be useful convenience functions when getting started on a new deep learning computer vision project or when you need to inspect specific images.
Some of these functions are demonstrated when working with pre-trained models in the Applications section of the API documentation.
All image handling in Keras requires that the Pillow library is installed.
If it is not installed, you can review the installation instructions.
Let’s take a closer look at each of these functions in turn.
Keras provides the load_img() function for loading an image from file as a PIL image object.
The example below loads the Bondi Beach photograph from file as a PIL image and reports details about the loaded image.
Running the example loads the image and reports details about the loaded image.
We can confirm that the image was loaded as a PIL image in JPEG format with RGB channels and the size of 640 by 427 pixels.
The loaded image is then displayed using the default application on the workstation, in this case, the Preview application on macOS.
Example of Displaying a PIL image using the Default ApplicationThe load_img() function provides additional arguments that may be useful when loading the image, such as ‘grayscale‘ that allows the image to be loaded in grayscale (defaults to False), ‘color_mode‘ that allows the image mode or channel format to be specified (defaults to rgb), and ‘target_size‘ that allows a tuple of (height, width) to be specified, resizing the image automatically after being loaded.
Keras provides the img_to_array() function for converting a loaded image in PIL format into a NumPy array for use with deep learning models.
The API also provides the array_to_img() function that can be used for converting a NumPy array of pixel data into a PIL image.
This can be useful if the pixel data is modified while the image is in array format and can then be saved or viewed.
The example below loads the test image, converts it to a NumPy array, and then converts it back into a PIL image.
Running the example first loads the photograph in PIL format, then converts the image to a NumPy array and reports the data type and shape.
We can see that the pixel values are converted from unsigned integers to 32-bit floating point values, and in this case, converted to the array format [height, width, channels].
Finally, the image is converted back into PIL format.
The Keras API also provides the save_img() function to save an image to file.
The function takes the path to save the image, and the image data in NumPy array format.
The file format is inferred from the filename, but can also be specified via the ‘file_format‘ argument.
This can be useful if you have manipulated image pixel data, such as scaling, and wish to save the image for later use.
The example below loads the photograph image in grayscale format, converts it to a NumPy array, and saves it to a new file name.
Running the example first loads the image and forces the format to be grayscale.
The image is then converted to a NumPy array and saved to the new filename ‘bondi_beach_grayscale.
jpg‘ in the current working directory.
To confirm that the file was saved correctly, it is loaded again as a PIL image and details of the image are reported.
The loaded grayscale image is then displayed using the default image preview application on the workstation, which in macOS is the Preview application.
Example of Saved Grayscale Image Shown Using the Default Image Viewing ApplicationThis section provides more resources on the topic if you are looking to go deeper.
In this tutorial, you discovered how to use the basic image handling functions provided by the Keras API.
Specifically, you learned:Do you have any questions?.Ask your questions in the comments below and I will do my best to answer.
.. More details