Non-Local Means (NLM) algorithm is widely considered as a state-of-the-art denoising filter in many research fields. High computational complexity led to implementations on Graphic Processor Unit (GPU) architectures, which achieve reasonable running times by filtering, slice-by-slice, 3D datasets with a 2D NLM approach. Here we present a fully 3D NLM implementation on a multi-GPU architecture and suggest its high scalability. The performance results we discuss encourage the coding of further filter improvements and the investigation of a large spectrum of applicative scenarios.
3D Non-Local Means denoising via multi-GPU / G., Palma; M., Comerci; B., Alfano; Cuomo, Salvatore; DE MICHELE, Pasquale; Piccialli, Francesco; Borrelli, Pasquale. - ELETTRONICO. - Computer Science and Information Systems (FedCSIS), 2013 Federated Conference:(2013), pp. 495-498. (Intervento presentato al convegno —6 th Computer Aspects of Numerical Algorithms tenutosi a Carocia, Polonia nel 2013).
3D Non-Local Means denoising via multi-GPU
CUOMO, SALVATORE;DE MICHELE, PASQUALE;PICCIALLI, FRANCESCO;BORRELLI, PASQUALE
2013
Abstract
Non-Local Means (NLM) algorithm is widely considered as a state-of-the-art denoising filter in many research fields. High computational complexity led to implementations on Graphic Processor Unit (GPU) architectures, which achieve reasonable running times by filtering, slice-by-slice, 3D datasets with a 2D NLM approach. Here we present a fully 3D NLM implementation on a multi-GPU architecture and suggest its high scalability. The performance results we discuss encourage the coding of further filter improvements and the investigation of a large spectrum of applicative scenarios.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
