Automated image analysis pipeline for scienfitic data, designed for reproducibility, modularity, and compatibility with Bernhard Lab instrumentation.
This repository contains Python modules, configuration files, and notebooks that work together to process and analyze images in batch. Core processing logic lives in Image_Processing_Suite, while your raw images and binary masks reside in a separate project folder following a specific layout.
Image_Processing_Suite/
│
├── image_processing/ # Core Python modules
├── config.yaml # Analysis parameters (thresholds, filters, etc.)
└── process_images.ipynb # Notebook to run the full pipeline
Create a project folder anywhere, with this structure:
<Your_Project_Folder>/
│
├── input/
│ ├── images/
│ │ ├── <channel1_name>/
│ │ └── <channel2_name>/
│ └── masks/
│ ├── <channel1_name>/
│ └── <channel2_name>/
│
└── inflection_finder.ipynb # Notebook for downstream data inspection
Place images and masks in their respective folders. These names should match the channel names in config.yaml. Ensure each channel folder under images has a matching folder under masks.
The inflection_finder.ipynb notebook is provided in example project folders (e.g., Ninetales_Template/) for further data inspection and analysis after processing. It will find and annotate inflection points in the processed data.