DistanceTransforms

DistanceTransforms.jl is a Julia package that provides efficient distance transform operations on arrays (CPU & GPU).

Installation

using Pkg
Pkg.add("DistanceTransforms")

Quick Start

using DistanceTransforms

# Create a simple binary array
arr = [
    0 1 1 0
    0 0 0 0
    1 1 0 0
]

# Apply distance transform
result = transform(boolean_indicator(arr))

Features

• Fast CPU implementation using the Felzenszwalb algorithm
• Multi-threading support for enhanced CPU performance
• GPU acceleration for NVIDIA (CUDA), AMD (ROCm), and Apple (Metal)
• Simple API with unified transform interface
• Multi-dimensional support for 1D, 2D, and 3D arrays

Advanced Usage

Multi-threading

# Compare single vs multi-threaded performance
result = transform(arr; threaded = true)  # default
result = transform(arr; threaded = false) # single-threaded

GPU Acceleration

using CUDA
# Automatically uses GPU implementation for CUDA arrays
x_gpu = CUDA.CuArray(boolean_indicator(rand([0, 1], 100, 100)))
gpu_result = transform(x_gpu)

Python Support

For Python users, check out distance_transforms, a Python wrapper providing the same functionality.

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)

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}}

Documentation

For comprehensive documentation and examples, visit our documentation site.