Title :
Parallel DSP architecture for reconstruction of tomographic images using wavelets techniques
Author :
Cruvinel, Paulo E.
Abstract :
This paper describes a technique for reconstructing tomographic images generated by a minitomograph scanner for soil science developed by Embrapa Agricultural Instrumentation using a parallel DSP architecture with an HEPC2E board from Hunt Engineering with two TMS32OC40 processors from Texas Instruments. The software designed for visualization of bi- and tri-dimensional images uses code composer tool combined with Parallel C from 3L and Borland Builder C++. Reconstruction and interpolation tasks were distributed throughout the platform processors, using parallel algorithms in order to improve speed and processing power. The work resulted in an increase of about 15% in the system speed when compared with that reached using a classical platform at 200 MHz
Keywords :
digital signal processing chips; geophysical techniques; image reconstruction; parallel algorithms; parallel architectures; seismology; wavelet transforms; Borland Builder C; HEPC2E board; Parallel C; TMS320C40 processors; code composer tool; images visualization; interpolation; minitomograph scanner; parallel DSP architecture; parallel algorithms; soil science; tomographic images reconstruction; wavelets techniques; Agricultural engineering; Computer architecture; Digital signal processing; Image generation; Image reconstruction; Instruments; Software design; Software tools; Soil; Tomography;
Conference_Titel :
Computer Graphics and Image Processing, 2001 Proceedings of XIV Brazilian Symposium on
Conference_Location :
Florianopolis
Print_ISBN :
0-7695-1330-1
DOI :
10.1109/SIBGRAPI.2001.963090
