barc4dip

barc4dip is a Python package for digital image processing, with a focus on scientific images and beam diagnostics.

It provides utilities for image I/O, preprocessing, signal analysis, plotting, and compact reporting, with current emphasis on speckle and sharpness metrics for 2D images and image stacks.

Features

• Read and write scientific images (TIFF, EDF, HDF5)
• Compute image statistics and distribution moments
• Compute speckle metrics on single images and stacks
• Compute sharpness metrics on single images and stacks
• Apply basic preprocessing tools such as flat-field correction and PSF deconvolution
• Perform FFT / PSD / correlation / translation tracking operations
• Plot images, histograms, tile maps, stack metrics, and displacements
• Export compact Markdown logbook reports
• Run a small CLI for speckle-image reporting

Installation

Local installation from source:

git clone https://github.com/barc4/barc4dip.git
cd barc4dip
pip install -e .

pip install barc4dip is planned, but not available yet.

Main modules

• barc4dip.io – image reading and writing
• barc4dip.preprocessing – flat-field correction, deconvolution
• barc4dip.metrics – speckle, sharpness, and distribution metrics
• barc4dip.signal – FFT, PSD, correlation, and tracking helpers
• barc4dip.plotting – plotting API for images and stack metrics
• barc4dip.report – Markdown logbook summaries
• barc4dip.geometry / maths / utils – supporting tools

Quick example

import barc4dip as dip

img = dip.read_image("image.tif")
stats = dip.sharpness_stats(img)
report = dip.logbook_report(stats)

print(stats["full"])
print(report)

For stacks:

import barc4dip as dip

stack = dip.read_image("scan.h5")
stats = dip.sharpness_stack_stats(stack)

Command line

A small CLI is currently available for speckle-image analysis:

barc4dip-speckles -s speckles.tif -o report.md

Optional flat and dark images can also be provided.

Status

This project is under active development.

• PyPI distribution: coming soon
• More examples: coming soon
• Full documentation: coming soon

The repository already contains early notebooks and generated report examples that illustrate the intended workflow.

License

CeCILL-2.1