Title :
A processor architecture for the image and volume reconstruction
Author :
Caquineau, C. ; Amans, J.
Author_Institution :
CEA-IRDI, CENG, Grenoble, France
Abstract :
The results of a study of the algorithms for image (2-D) and volume (3-D) reconstruction from projections are presented. The algorithms are studied to extract a common structure in order to allow design of suitable processor architectures for this task. The different algorithms and their inner operations are presented and a possible computing module for 3-D reconstruction is discussed. The implementation of the algorithms in a parallel machine including such processors is outlined. The image-reconstruction algorithms are characterized by an algorithmic complexity of high order. The module is characterized by high performance and a relative versatility, and it allows a certain choice not only in the geometry but also in the reconstruction methodology
Keywords :
computerised picture processing; parallel algorithms; parallel architectures; algorithmic complexity; computing module; geometry; image-reconstruction algorithms; inner operations; parallel machine; processor architecture; reconstruction methodology; volume reconstruction; Algorithm design and analysis; Apertures; Computer architecture; Geometry; Image reconstruction; Iterative algorithms; Parallel machines; Parallel processing; Partitioning algorithms; Process design; Radiography; Reconstruction algorithms; Three dimensional displays;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1990. ICASSP-90., 1990 International Conference on
Conference_Location :
Albuquerque, NM
DOI :
10.1109/ICASSP.1990.115852
