FDmap - Fractal Dimension maps for analyzing restored paintings

Juan Ruiz de Miras. University of Granada - SPAIN. 2026.

Last update: March - 2026

The fractal dimension (FD) is a quantitative measure of complexity that has been effectively used to analyze paintings. FDmap is a novel methodology that combines FD computation on color images with a sliding window approach to generate detailed FD maps of paintings before and after restoration.

You can download the C++/CUDA/MATLAB source code of FDmap by clicking here (138 MB).
Example dataset can be downloaded by clicking here (80 MB).

Once unzipped the file fdmap.zip, you will find the following folders and files:

• Folder code/, contains the source code of FDmap, organized as:
  • 1_Maps_Creation/, folder with the C++/CUDA code for generating the FD maps:
    • fdmap.cpp and fdmap.h: C++ files for generating FD maps
    • test.cpp: example of C++ main file generating in parallel the FD maps for the dataset of paintings
    • external/: folder with external code for DBC-RGB computing (by Juan Ruiz de Miras), linear regression computation (by geeksforgeeks.org) and reading BMP images (by Arash Partow)
  • 2_Maps_Visualization/, folder with the MATLAB code for displaying FD maps and saving as .bmp files
    • FDmapVisualization.m: MATLAB code for displaying the FD maps and saving them as .png images
  • license.html: terms of the license for the source code of FDmap
  • README.txt: text file containing the present information about the program
• dataset/: empty folder where the file dataset.zip must be unzipped
• results/: folder where generated FD maps and .png images will be saved. It contains pre-computed FD maps and .png images for the dataset.

REQUERIMENTS for compilation and execution

Hardware: Computer with a CUDA capable 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)
• MATLAB (https://www.mathworks.com/products/matlab.html)

COMPILATION AND RUNNING

1. Go to folder code/1_Maps_Creation/:
   • The test.cpp file can be compiled with the following command (change MATLAB_HOME with the path of your local installation of MATLAB):
     • Windows: nvcc -Xcompiler "/openmp" -I"MATLAB_HOME\extern\include" -L"MATLAB_HOME\extern\lib\win64\microsoft" -llibmat -llibmex -llibmx -O3 -o test external/dbc-rgb.cu fdmap.cpp test.cpp
     • Linux: nvcc -Xcompiler "-fopenmp" -IMATLAB_HOME/extern/include -LMATLAB_HOME/bin/glnxa64 -lmat -lmex -lmx -O3 -o test external/dbc-rgb.cu fdmap.cpp test.cpp
   • Run the file test to process the dataset. Results with the .mat files of the FD maps for each painting are saved in the folder results
2. Go to folder code/2_Maps_Visualization/:
   • Run the MATLAB file FDmapVisualization.m to process the .mat files generated in the previous step. The FD maps for all paintings will be displayed and saved as .png images in the folder results

CITATION

CONTACT AND SUGGESTIONS

FUNDING