ImageIO

Language: Python

Data Science

ImageIO was developed to unify the reading and writing of images and videos in Python with a simple and consistent API. It is widely used in scientific computing, machine learning, and multimedia applications for processing image and video data.

ImageIO is a Python library that provides an easy interface to read and write images in a wide range of formats, including PNG, JPEG, BMP, GIF, TIFF, and more. It also supports reading and writing video files and volumetric data.

Installation

pip: pip install imageio

conda: conda install -c conda-forge imageio

Usage

ImageIO allows you to read, write, and process images and videos using NumPy arrays. It provides a simple API to load images into arrays, perform operations, and save results. Plugins are available for handling different formats and compression options.

Reading an image

import imageio
img = imageio.imread('example.png')
print(img.shape)

Reads an image from a file and loads it as a NumPy array.

Writing an image

import imageio
imageio.imwrite('output.png', img)

Saves a NumPy array as an image file.

Reading a video

import imageio
reader = imageio.get_reader('video.mp4')
for frame in reader:
    print(frame.shape)
reader.close()

Reads frames from a video file and prints their dimensions.

Writing a video

import imageio
writer = imageio.get_writer('output.mp4', fps=24)
for frame in frames:  # frames is a list of NumPy arrays
    writer.append_data(frame)
writer.close()

Writes a series of frames (NumPy arrays) to a video file at 24 FPS.

Reading images from URLs

import imageio
img = imageio.imread('https://example.com/image.jpg')

Directly reads an image from a web URL into a NumPy array.

Using plugins for format-specific options

import imageio
img = imageio.imread('example.tiff', format='TIFF')

Specifies a plugin format explicitly when reading an image to handle special format options.

Error Handling

FileNotFoundError: Ensure the file path exists and is accessible.

ValueError: Cannot identify image file: Check that the file format is supported or specify the format explicitly.

RuntimeError: Cannot write frames: Ensure that the writer is opened properly and frames are valid NumPy arrays with correct shape and dtype.

Best Practices

Use NumPy arrays for all image and video processing for efficiency.

Close readers and writers properly to free resources.

Specify formats explicitly when needed to avoid ambiguities.

Use plugins for advanced format handling (e.g., TIFF, GIF, DICOM).

Combine ImageIO with libraries like NumPy, OpenCV, or PIL for processing pipelines.

Official Docs Github