Fast Box-Counting Algorithm

Juan Ruiz de Miras and Miguel Ángel Posadas. University of Granada - SPAIN. 2022.

Last update: october - 2022

The box-counting (BC) algorithm is one of the most popular methods for calculating the fractal dimension (FD) of binary data. However, high computation times are needed to calculate the BC of large datasets composed of data in 2D, 3D or 4D. Here, we provide the source code of a very efficient parallel implementation of the BC algorithm for its execution on Graphics Processing Units (GPU).

You can download the C++/CUDA source code of the Fast Box-Counting algorithm by clicking here.

The file fbc.zip contains:

• bcCUDA2D.cu and bcCUDA2D.cuh: CUDA/C++ files with the implementation on GPU of the box-counting algorithm for binary 2D data
• bcCUDA3D.cu and bcCUDA3D.cuh: CUDA/C++ files with the implementation on GPU of the box-counting algorithm for binary 3D data
• bcCUDA4D.cu and bcCUDA4D.cuh: CUDA/C++ files with the implementation on GPU of the box-counting algorithm for binary 4D data
• bcCPU.cpp and bcCPU.h: C++ files with the implementation on CPU of the box-counting algorithm for binary 2D, 3D and 4D data
• test.cpp: example of C++ main file executing the GPU box-counting algorithms
• license.html: terms of the license for the source code of Fast Box-counting
• README.txt: text file containing the present information about the program

REQUERIMENTS for compilation and execution

Hardware: Computer with a capable CUDA GPU (https://developer.nvidia.com/cuda-gpus)
Software:

• CUDA SDK Toolkit (https://developer.nvidia.com/cuda-toolkit)
• C++ compiler (the binaries folder must be included in the enviroment variable PATH of the OS)

COMPILATION

The test.cpp file can be compiled with the following command: > nvcc -O3 -o test bcCUDA4D.cu bcCUDA3D.cu bcCUDA2D.cu bcCPU.cpp test.cpp

CITATION

CONTACT AND SUGGESTIONS