In this paper, we present a new parallel implementation of the Vertex Component Analysis (VCA) algorithm for spectral unmixing of remotely sensed hyperspectral data on commodity graphic processing units (GPUs). We first developed a C serial version of the VCA algorithm and three parallel versions: one using NVidia’s Compute Unified Device Architecture (CUDA), another using CUDA basic linear algebra subroutines (CUBLAS) library and the last using the CUDA linear algebra (CULA) library. Experimental results, based on the analysis of hyperspectral images acquired by a variety of hyperspectral imaging sensors, show the effectiveness of our implementation, which satisfies the real-time constraints given by the data acquisition rate.
Real-Time Implementation of the Vertex Component Analysis Algorithm on GPUs
BARBERIS, ALESSANDRO;DANESE, GIOVANNI;LEPORATI, FRANCESCO;TORTI, EMANUELE
2012-01-01
Abstract
In this paper, we present a new parallel implementation of the Vertex Component Analysis (VCA) algorithm for spectral unmixing of remotely sensed hyperspectral data on commodity graphic processing units (GPUs). We first developed a C serial version of the VCA algorithm and three parallel versions: one using NVidia’s Compute Unified Device Architecture (CUDA), another using CUDA basic linear algebra subroutines (CUBLAS) library and the last using the CUDA linear algebra (CULA) library. Experimental results, based on the analysis of hyperspectral images acquired by a variety of hyperspectral imaging sensors, show the effectiveness of our implementation, which satisfies the real-time constraints given by the data acquisition rate.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
