Time management and synchronization for real-time image processing using multiple cameras and PC cluster.

Accession number;02A0234232
Title;Time management and synchronization for real-time image processing using multiple cameras and PC cluster.
Author; ARITA DAISAKU (Kyushu Univ., Graduate School of Informaiton Sci. and Electrical Engineering, JPN) TANIGUCHI RIN'ICHIRO (Kyushu Univ., Graduate School of Informaiton Sci. and Electrical Engineering, JPN)
Journal Title;Joho Shori Gakkai Kenkyu Hokoku
Journal Code:Z0031B
ISSN:0919-6072
VOL.2002;NO.2(CVIM-131);PAGE.9-16(2002)
Figure&Table&Reference;FIG.8, REF.5
Pub. Country;Japan
Language;Japanese
Abstract;For real-time image processing using multiple cameras and PC cluster, all cameras and PCs must share a clock. And for continuous real-time image processing, data processing and data transfer must be synchronized between all PCs. To solve these problems, we are construct RPV, a programming environment. RPV, which assumes that all cameras are synchronized by external synchronization signal, have a mechanism to synchronize all cameras and PCs by sharing the synchronization signal with all PCs. Using RPV, an application programmer can easily develop real-time image processing systems by describing only data flow information between PCs and data processing tasks. (author abst.)