This code package implements the image generation methodology outlined in:
Rajendran, L. K., Bane, S., & Vlachos, P. (2019). PIV/BOS synthetic image generation in variable density environments for error analysis and experiment design. Measurement Science and Technology.
Please cite the above paper if you use this code package for your work.
UPDATED Instructions for running the software.
- Navigate to cuda_codes/Debug/
- Run "makefile_simplified" in the terminal - this should compile the codes and generate the file "libparallel_ray_tracing.so"
- Run "ls -lt libparallel_ray_tracing.so" and check the timestamp to ensure that the file has been recently updated.
- Type "export LD_LIBRARY_PATH=$(cd ../ && pwd)/lib:$(cd ../ && pwd)/lib64" in the terminal command
- Then type "echo $LD_LIBRARY_PATH" and confirm that it displays something like: "path-to-photon/cuda_codes/lib:path-to-photon/cuda_codes/lib64"
- To avoid performing the above two steps everytime, add the two commands to a bash initialization script (such as .bashrc or .profile)
- Navigate to python_codes
- Run "sample_run_script.sh piv" to run a sample piv simulation or "sample_run_script.sh bos" to run a sample bos simulation
Dependencies for Python:
- numpy
- scipy
- matplotlib
- libtiff
- ctypes
Dependencies for CUDA:
- bz2 (https://www.sourceware.org/bzip2/)
- png (http://www.libpng.org/pub/png/libpng.html)
- teem (http://teem.sourceforge.net)
- cubic_interpolation_cuda (https://github.com/DannyRuijters/CubicInterpolationCUDA)
Sample data is provided in the sample-data/ directory. Run the "sample_run_script.sh" file after compiling the software to run the package and generate sample images. You will have to make the bash script executable before running it. To do that, type: "chmod +x sample_run_script.sh" in the terminal.
The script light_ray_processing.py contains useful python functions to load, manipulate, and display light ray data.