MIMOSA (Microscopy IMage prOcessing SoftwAre) is a node-based image analysis toolkit for accurate dispersion measurements for large datasets of SEM images. It provides a modular framework for particle segmentation and dispersion measurement, optimized for large image batches and research on rubber–lignin composites. This software was developed alongside a Master's thesis, which can be found here.
- Node-based visual pipeline for image processing
- High-performance backend using NumPy, SciPy, OpenCV, scikit-image, and Numba
- Real-time GUI built with Pygame
- Builtin support for TensorFlow semantic segmentation models
- Tools for thresholding, segmentation, and morphological analysis
- Designed for SEM images and particle dispersion studies
git clone https://github.com/dille12/MIMOSA.git
cd MIMOSA
pip install -r requirements.txtOn Linux, TKinter must be installed via bash:
sudo apt-get update
sudo apt-get install python3-tk
Download the neural network package from the Releases tab by clicking here.