مقاله GPU-accelerated level-set segmentation : تی پی بین

خلاصه مقاله

The level-set method, a technique for the computation of evolving interfaces, is a solution commonly used to segment images and volumes in medical applications. GPUs have become a commodity hardware with hundreds of cores that can execute thousands of threads in parallel, and they are nowadays ideal platforms to execute computational intensive tasks, such as the 3D level-set-based segmentation, in real time. In this paper, we propose two GPU implementations of the level-set-based segmentation method called Fast Two-Cycle. Our proposals perform computations in independent domains called tiles and modify the structure of the original algorithm to better exploit the features of the GPU. The implementations were tested with real images of brain vessels and a synthetic MRI image of the brain. Results show that they execute faster than a CPU-sequential implementation of the same method, without any significant loss of the segmentation quality and without requiring distributed parallel computer infrastructures.

نویسندگان

Julián Lamas-Rodríguez [ Centro de Investigación en Tecnoloxías da Información da Universidade de Santiago de Compostela (CITIUS), Santiago de Compostela, Spain ]

Dora B. Heras [ Centro de Investigación en Tecnoloxías da Información da Universidade de Santiago de Compostela (CITIUS), Santiago de Compostela, Spain ]

Francisco Argüello [ Department of Electronics and Computer Science, University of Santiago de Compostela, Santiago de Compostela, Spain ]

Dagmar Kainmueller [ Department of Visualization and Data Analysis, Konrad-Zuse-Zentrum für Informationstechnik, Berlin, Germany ]

Stefan Zachow [ Department of Visualization and Data Analysis, Konrad-Zuse-Zentrum für Informationstechnik, Berlin, Germany ]

کلمات کلیدی

Level set Segmentation GPU CUDA GPGPU

آدرس این مقاله

آدرس مستقیم دسترسی به این مقاله

مشخصات کنفرانس

مشاهده مقالات

Journal of Real-Time Image Processing 2013

Although there are many journals addressing the subject of image processing, the Journal of Real-Time Image Processing (JRTIP) is the only one that is solely dedicated to the real-time aspect of image and video processing. It is often the case that an image processing algorithm is developed and proven theoretically sound, presumably with a specific application in mind, but its practical applications and the detailed steps, methodology, and trade-off analysis required to achieve its real-time performance are never fully explored. JRTIP is thus intended to bridge the gap between the theory and practice of image and video processing, serving the greater community of researchers, practicing engineers, and industrial professionals who deal with designing, implementing or utilizing real-time image and video processing systems. The real-time aspect is critical in many real-world devices or products such as mobile phones, digital still/video/cell-phone cameras, portable media players, personal digital assistants, high-definition television, video surveillance systems, industrial visual inspection systems, medical imaging devices, vision-assisted intelligent robots, spectral imaging systems, and many other embedded image or video processing systems. Topics covered in the journal include: Real-time image and video processing algorithmsReal-time embedded image/video processing systems Real-time image and video processing hardware and architecture including FPGA, DSP, GPU, GPP, ASIC, System-on-Chip (SoC), and System-in-a-Package (SiP) implementations Real-time software optimizations and related design paradigms for image/video processingReal-time hardware/software co-design for image and video processing Real-time image and video processing applications including digital, cell-phone, and smart cameras, machine vision, industrial inspection, surveillance and security, image and video compression, for transmission and for database storage and retrieval, biomedical imaging, spectral imaging, etc.

حوزه های تحت پوشش : Image Processing and Computer Vision, Multimedia Information Systems, Computer Graphics, Pattern Recognition