image-stitcher

image stitching for microscopy with GPU acceleration support.

Setup from Scratch

Ubuntu 22.04

./setup_ubuntu_22_04.sh

Installs conda (if needed), creates image-stitcher environment, and installs dependencies.

Ubuntu 22.04 with CUDA (Optional)

./setup_cuda_22.04.sh

Adds CuPy for optimal NVIDIA GPU performance (compatible with Driver 535.x).

Windows 11

Note: If you haven't enabled script execution in PowerShell, you may need to run this first:

Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser

Then run the setup:

.\setup_windows_11.ps1

Installs conda (if needed), creates image-stitcher environment, and installs dependencies.

Windows 11 with CUDA (Optional)

.\setup_cuda_windows_11.ps1

Adds CuPy for optimal NVIDIA GPU performance.

Desktop Shortcut (Optional)

python create_desktop_shortcut.py

Creates a desktop launcher that auto-activates the conda environment.

Launch

GUI

./run_gui

Or manually:

conda activate image-stitcher
python -m image_stitcher.stitcher_gui

GPU Acceleration in GUI

The GUI includes a Compute Backend dropdown that allows you to select the tensor backend:

• Auto (Recommended) - Automatically selects the best available backend
• NumPy (CPU) - CPU-only processing using NumPy
• PyTorch (GPU/CPU) - Uses PyTorch with GPU acceleration if available
• CuPy (NVIDIA GPU) - Maximum performance with NVIDIA GPU acceleration

The dropdown shows the actual backend status (GPU/CPU).. Backend selection affects the registration process performance significantly.

Devtools

This repository is set up with ruff for linting and formatting, and mypy for type checking. The shell scripts in the dev directory can be used to invoke these tools and should be run from the repository root.

You can install them with pip install mypy ruff