DeepTrack2
An MIT-licensed Python library for generating and processing image data for machine learning and digital microscopy. Current versions don't support TensorFlow.
DeepTrack2 is an open-source Python library for researchers building image data pipelines for machine learning and digital microscopy. It brings image generation, manipulation and analysis into a modular framework that runs in your own Python environment. Its focus is scientific imaging, with applications in microscopy, optics and biophysics.
The library covers data augmentation and image processing alongside optical components for holography, aberrations, noise and scatterers. These capabilities make it relevant to projects that need to generate or modify microscopy images as part of their training data. It also includes tools for image sequences and statistics, plus sources for loading images from folders.
DeepTrack2 provides introductory notebooks and examples that connect these capabilities to model training. One example uses a conditional GAN to generate realistic electron microscopy images; another trains a fully connected neural network to recognize handwritten digits in MNIST. The examples give prospective users concrete tasks to compare with their own research needs.
The library uses the MIT license and requires Python 3.10 or later. Its current code doesn't support TensorFlow.
Logotyper
