GPU-optimized Direct Fourier Method for On-line Tomography / A. V. Shkarin [et al.]

Уровень набора: Fundamenta InformaticaeАльтернативный автор-лицо: Шкарин, А. В., специалист в области информатики и вычислительной техники, учебный мастер Томского политехнического университета, 1990-, Андрей Владимирович;Ametova, E. S., specialist in the field of automation and computer systems, assistant at Tomsk Polytechnic University, 1988-, Evelina Serverovna;Chilingaryan, S., Suren;Dritschler, T., Timo;Kopmann, A., Andreas;Vogelgesang, M., Matthias;Shkarin, R. V., specialist in the field of informatics and computer technology, technician of Tomsk Polytechnic University, 1990-, Roman Vladimirovich;Tsapko, S. G., specialist in the field of informatics and computer technology, Associate Professor of Tomsk Polytechnic University, Head of laboratory, Candidate of technical sciences, 1973-, Sergey Gennadyevich;Mirone, A., AlessandroКоллективный автор (вторичный): Национальный исследовательский Томский политехнический университет (ТПУ), Институт кибернетики (ИК), Кафедра автоматики и компьютерных систем (АИКС)Язык: английский.Страна: Россия.Резюме или реферат: On-line monitoring of synchrotron 3D-imaging experiments requires very fast tomographic reconstruction. Direct Fourier methods (DFM) have the potential to be faster than standard Filtered Backprojection. We have evaluated multiple DFMs using various interpolation techniques. We compared reconstruction quality and studied the parallelization potential. A method using Direct Fourier Inversion (DFI) and a sinc-based interpolation was selected and parallelized for execution on GPUs. Several optimization steps were considered to boost the performance. Finally we evaluated the achieved performance for the latest generation of GPUs from NVIDIA and AMD. The results show that tomographic reconstruction with a throughput of more than 1.5 GB/sec on a single GPU is possible..Аудитория: .Тематика: электронный ресурс | труды учёных ТПУ | компьютерная томография | томографическая реконструкция Ресурсы он-лайн:Щелкните здесь для доступа в онлайн
Тэги из этой библиотеки: Нет тэгов из этой библиотеки для этого заглавия. Авторизуйтесь, чтобы добавить теги.
Оценка
    Средний рейтинг: 0.0 (0 голосов)
Нет реальных экземпляров для этой записи

Title screen

On-line monitoring of synchrotron 3D-imaging experiments requires very fast tomographic reconstruction. Direct Fourier methods (DFM) have the potential to be faster than standard Filtered Backprojection. We have evaluated multiple DFMs using various interpolation techniques. We compared reconstruction quality and studied the parallelization potential. A method using Direct Fourier Inversion (DFI) and a sinc-based interpolation was selected and parallelized for execution on GPUs. Several optimization steps were considered to boost the performance. Finally we evaluated the achieved performance for the latest generation of GPUs from NVIDIA and AMD. The results show that tomographic reconstruction with a throughput of more than 1.5 GB/sec on a single GPU is possible.

Для данного заглавия нет комментариев.

оставить комментарий.