Image processing 4 Dummies
FREE, USER-FRIENDLY PROTOCOLS for image processing
We understand that as a biologist, it can be challenging to access and fully utilize the cutting-edge tools developed by computational scientists. While many of these tools come with detailed descriptions, we've created simple, step-by-step guides to help you navigate some of the most advanced image processing tools with ease. Don’t forget to explore the other tools listed below as well!
Micro-SAM (μSAM) is a tool designed for interactive segmentation and tracking in microscopy images, built on top of the Segment Anything model. It enables accurate segmentation in 2D and 3D for diverse microscopy modalities, including light and electron microscopy. Through a Napari plugin, it provides both interactive and automatic segmentation, enhancing the efficiency and quality of image analysis across various microscopy tasks.

Ilastik is an easy-to-use image analysis tool that uses machine learning to segment, classify, track, and count cells or other experimental data. You simply draw labels on the image, and the software learns from your input to classify pixels or objects. Most operations are interactive, allowing you to see results immediately, even on large datasets—no machine learning expertise required!

Batch Training Pixel Classifiers in Imaris with the Frankenstein Script
The Frankenstein script enables the training of a pixel classifier within Imaris (Oxford Instruments) using multiple image files simultaneously. Instead of training on a single dataset, this script extracts timepoints from all .ims files in a specified directory and compiles them into a new file. This new dataset consists of randomly selected sections from each original file, allowing for a more diverse and representative training set for the pixel classifier.

StarDist is a tool designed for cell/nuclei segmentation in microscopy images. It is a deep learning based 2D and 3D object detection method with star-convex shapes. StarDist has originally been developed for the segmentation of densely packed cell nuclei in challenging images with low signal-to-noise ratios. A napari plugin for StarDist allows to apply pretrained and custom trained models from within napari.
