Distance Transforms

DistanceTransforms.jl and distance_transforms provides efficient, GPU-accelerated, distance transform operations for arrays in both Julia and Python. Distance transforms are essential in computer vision, image processing, and machine learning tasks.

Example of a distance transform applied to various shapes

Features

Feature	Description
🚀 High Performance	Optimized algorithms for CPU and GPU
🧵 Multi-threading	Parallel processing on CPU
🖥️ GPU Acceleration	Support for NVIDIA (CUDA), AMD (ROCm), Apple (Metal), and Intel (oneAPI)
🐍 Python Integration	Full Python support via `distance_transforms`
📏 Versatile Dimensions	Works with 1D, 2D, and 3D arrays
📚 Well Documented	Comprehensive guides and examples

Acknowledgements

This project would not be possible without several excellent Julia packages:

KernelAbstractions.jl: Provides hardware-agnostic GPU code that runs across different GPU platforms
AcceleratedKernels.jl: Makes it dead simple to write GPU kernels with normal Julia code
PythonCall.jl: Powers the Python interoperability layer for seamless integration
DLPack.jl: Enables zero-copy tensor sharing with PyTorch for efficient GPU operations

Why This Library?

	Distance Transforms	ImageMorphology.jl	SciPy
Fast Distance Transform	✅✅	✅	✅
CPU Single-Threaded Support	✅	✅	✅
CPU Multi-Threaded Support	✅	✅	❌
NVIDIA/CUDA Support	✅	❌	❌
AMD/ROCM Support	✅	❌	❌
Apple/Metal Support	✅	❌	❌
Intel/oneAPI Support	✅	❌	❌
Python Integration	✅	❌	✅

Choose Your Language

Julia

using DistanceTransforms
arr = rand([0, 1], 10, 10)
result = transform(boolean_indicator(arr))

Get started with Julia →

Python

import numpy as np
import distance_transforms as dts
arr = np.random.choice([0, 1], size=(10, 10)).astype(np.float32)
result = dts.transform(arr)

Get started with Python →

Installation

Julia

using Pkg
Pkg.add("DistanceTransforms")

Python

pip install distance_transforms

Citation

If you use DistanceTransforms in your research, please cite:

@ARTICLE{10912438,
  author={Black, Dale and Li, Wenbo and Zhang, Qiyu and Molloi, Sabee},
  journal={IEEE Access}, 
  title={Accelerating Euclidean Distance Transforms: A Fast and Flexible Approach with Multi-Vendor GPU, Multi-Threading, and Multi-Language Support}, 
  year={2025},
  volume={},
  number={},
  pages={1-1},
  keywords={Transforms;Graphics processing units;Benchmark testing;Euclidean distance;Three-dimensional displays;Image segmentation;Hardware;Python;Metals;Arrays;Distance Transform;Hardware Acceleration;Hausdorff Distance;High Performance Computing;Image Processing;Medical Image Analysis;Parallel Computing;Segmentation;Skeletonization},
  doi={10.1109/ACCESS.2025.3548563}}

License

DistanceTransforms is available under the MIT License.

--- title: "Distance Transforms" format: html: toc: true --- <style> .add-space-r{ padding-right: 2%; } .add-space-l{ padding-left: 2%; } </style> [`DistanceTransforms.jl`](julia/index.qmd) and [`distance_transforms`](python/index.qmd) provides efficient, GPU-accelerated, distance transform operations for arrays in both Julia and Python. Distance transforms are essential in computer vision, image processing, and machine learning tasks. ![Example of a distance transform applied to various shapes](images/various_dts.png) ## Features | Feature | Description | |---------|-------------| | 🚀 **High Performance** | Optimized algorithms for CPU and GPU | | 🧵 **Multi-threading** | Parallel processing on CPU | | 🖥️ **GPU Acceleration** | Support for NVIDIA (CUDA), AMD (ROCm), Apple (Metal), and Intel (oneAPI) | | 🐍 **Python Integration** | Full Python support via `distance_transforms` | | 📏 **Versatile Dimensions** | Works with 1D, 2D, and 3D arrays | | 📚 **Well Documented** | Comprehensive guides and examples | ## Acknowledgements This project would not be possible without several excellent Julia packages: - **KernelAbstractions.jl**: Provides hardware-agnostic GPU code that runs across different GPU platforms - **AcceleratedKernels.jl**: Makes it dead simple to write GPU kernels with normal Julia code - **PythonCall.jl**: Powers the Python interoperability layer for seamless integration - **DLPack.jl**: Enables zero-copy tensor sharing with PyTorch for efficient GPU operations ## Why This Library? | | Distance Transforms | ImageMorphology.jl | SciPy | |-----------------------|:---------------------:|:---------------:|:---------------:| | Fast Distance Transform | ✅✅ | ✅ | ✅ | | CPU Single-Threaded Support | ✅ | ✅ | ✅ | | CPU Multi-Threaded Support | ✅ | ✅ | ❌ | | NVIDIA/CUDA Support | ✅ | ❌ | ❌ | | AMD/ROCM Support | ✅ | ❌ | ❌ | | Apple/Metal Support | ✅ | ❌ | ❌ | | Intel/oneAPI Support | ✅ | ❌ | ❌ | | Python Integration | ✅ | ❌ | ✅ | ## Choose Your Language :::: {.columns} ::: {.column width="50%" .add-space-r} ### Julia ```julia using DistanceTransforms arr = rand([0, 1], 10, 10) result = transform(boolean_indicator(arr)) ``` [Get started with Julia →](julia/index.qmd) ::: ::: {.column width="50%" .add-space-l} ### Python ```python import numpy as np import distance_transforms as dts arr = np.random.choice([0, 1], size=(10, 10)).astype(np.float32) result = dts.transform(arr) ``` [Get started with Python →](python/index.qmd) ::: :::: ## Installation :::: {.columns} ::: {.column width="50%" .add-space-r} ### Julia ```julia using Pkg Pkg.add("DistanceTransforms") ``` ::: ::: {.column width="50%" .add-space-l} ### Python ```bash pip install distance_transforms ``` ::: :::: ## Citation If you use DistanceTransforms in your research, please cite: ```bibtex @ARTICLE{10912438, author={Black, Dale and Li, Wenbo and Zhang, Qiyu and Molloi, Sabee}, journal={IEEE Access}, title={Accelerating Euclidean Distance Transforms: A Fast and Flexible Approach with Multi-Vendor GPU, Multi-Threading, and Multi-Language Support}, year={2025}, volume={}, number={}, pages={1-1}, keywords={Transforms;Graphics processing units;Benchmark testing;Euclidean distance;Three-dimensional displays;Image segmentation;Hardware;Python;Metals;Arrays;Distance Transform;Hardware Acceleration;Hausdorff Distance;High Performance Computing;Image Processing;Medical Image Analysis;Parallel Computing;Segmentation;Skeletonization}, doi={10.1109/ACCESS.2025.3548563}} ``` ## License DistanceTransforms is available under the MIT License.