Hello everyone,
I am running a benchmark “run_benchmark_pet_histogram.sh”, downloaded from the Castor site. It generates two files, “benchmark_pet_histogram_challenger_it2.hdr” and “benchmark_pet_histogram_challenger_it2.img” and it compares the files generated with the reference files. I am copying the results below. Can anyone please explain me how can I open the two files generated and why the results of the benchmark PET are not overlapping with the reference files.
Check benchmark results (compare reconstructed image with the reference one)
Hi Kajal,
Both reference and output images are saved in interfile. You should be able to open it with any image editor supporting this format. The .img file is a .raw image of 112x112x109 voxels of 2.0x2.0x2.027 mm.
Regarding your results, several users with recent compilers experienced similar issues with the benchmark. The difference values weren’t as high as in your results though. Could you tell us what is your computer hardware/OS ? And what is your compiler ? Is there any artifact in the generated image ?
Did you try the other benchmarks as well ?
Best regards,
Thibaut
Nothing seems to be wrong in the slice presented in the screenshot.
Maybe the best would be to look at the image generated from the difference between the reference image and the image reconstructed by your system, and check where do the differences come from.